hermes-agent/locales/en.yaml
kshitijk4poor bdde674296 feat(fast): /fast ultrafast for GPT-6 Astra, priced at the served tier
OpenAI's Ultrafast service tier (service_tier="ultrafast") is broadly
available for GPT-6 Astra in the API and Codex (Pro 500 / Enterprise),
billed at 6x Standard. GPT-6.1 Sol Ultrafast is announced as coming soon.

- One tier table: agent.fast_mode.parse_service_tier / STATIC_TIERS now
  back the CLI, gateway and TUI config loaders (three hand-copied parsers).
- resolve_fast_mode_overrides(tier="ultrafast") sends the tier only for
  Ultrafast models on api.openai.com / chatgpt.com; anything else gets
  nothing rather than a silent swap to another paid tier.
- /fast ultrafast on CLI, gateway (typed + picker when supported) and TUI;
  static-tier attach sites pass the tier through.
- Pricing: the response's SERVED service_tier is folded into usage (Codex
  stream assembler keeps it); served ultrafast prices from a published
  6x row (272K whole-request tier included). A request asking for
  Ultrafast but served at default bills at Standard.
- i18n: 4 new keys in all 17 locales; option lists mention ultrafast.
2026-09-29 23:54:34 +05:30

3983 lines
246 KiB
YAML
Raw Permalink Blame History

This file contains invisible Unicode characters

This file contains invisible Unicode characters that are indistinguishable to humans but may be processed differently by a computer. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

approval:
dangerous_header: "⚠️ DANGEROUS COMMAND: {description}"
choose_long: " [o]nce | [s]ession | [a]lways | [d]eny"
choose_short: " [o]nce | [s]ession | [d]eny"
prompt_long: " Choice [o/s/a/D]: "
prompt_short: " Choice [o/s/D]: "
choose_smart_deny: " [o]nce | [d]eny"
prompt_smart_deny: " Choice [o/D]: "
smart_deny_once_inputs: "o,once"
smart_deny_deny_inputs: "d,deny"
timeout: " ⏱ No answer in {waited} — the command was not run. Ask again to retry, or raise the limit: hermes config set approvals.timeout {suggested}"
allowed_once: " ✓ Allowed once"
allowed_session: " ✓ Allowed for this session"
allowed_always: " ✓ Added to permanent allowlist"
denied: " ✗ Denied"
cancelled: " ✗ Cancelled"
blocklist_message: "This command is on the unconditional blocklist and cannot be approved."
inputs:
approve: "approve,yes,ok,okay,confirm,y"
deny: "deny,no,reject,cancel,n"
always: "always,approve always,always approve"
session: "session,approve session,session approve"
confirm_once: "approve,approve once,once,yes,ok,confirm"
confirm_always: "always,always approve,remember"
confirm_cancel: "cancel,nevermind,no,deny"
summary:
denied: "You denied this {noun} — it did not run."
timeout: "No answer within {window} — the {noun} did not run."
notify_failed: "The approval request could not be delivered — the {noun} did not run."
cancelled: "The approval prompt was withdrawn or never reached you — the {noun} did not run."
blocked: "This {noun} is not allowed in an unattended session — it did not run."
default: "This {noun} did not run."
noun:
command: "command"
code: "code"
action: "action"
# Approval timeout window (tools/approval_context.py format_approval_window): 300 → "5 minutes".
window:
hour_one: "{count} hour"
hour_other: "{count} hours"
minute_one: "{count} minute"
minute_other: "{count} minutes"
second_one: "{count} second"
second_other: "{count} seconds"
gateway:
approval_expired: "⚠️ Approval expired (agent is no longer waiting). Ask the agent to try again."
draining: "⏳ Draining {count} active agent(s) before restart..."
goal_cleared: "✓ Goal cleared."
no_active_goal: "No active goal."
config_read_failed: "⚠️ Could not read config.yaml: {error}"
config_save_failed: "⚠️ Could not save config: {error}"
model:
error_prefix: "Error: {error}"
switched: "Model switched to `{model}`"
provider_label: "Provider: {provider}"
context_label: "Context: {tokens} tokens"
max_output_label: "Max output: {tokens} tokens"
cost_label: "Cost: {cost}"
capabilities_label: "Capabilities: {capabilities}"
prompt_caching_enabled: "Prompt caching: enabled"
warning_prefix: "Warning: {warning}"
saved_global: "Saved to config.yaml (`--global`)"
session_only_hint: "_(session only — add `--global` to persist)_"
current_label: "Current: `{model}` on {provider}"
current_tag: " (current)"
more_models_suffix: " (+{count} more)"
usage_switch_model: "`/model <name>` — switch model"
usage_switch_provider: "`/model <name> --provider <slug>` — switch provider"
usage_persist: "`/model <name> --global` — persist"
err_skew: "This gateway is running code from {boot_rev} but the checkout on disk is now {disk_rev}. Switching models would risk a stale-module crash — restart the gateway to load the new code: hermes gateway restart"
err_switch_failed: "Model switch to {model} failed ({error}); staying on {current}."
err_config_not_updated: "config.yaml not updated ({error})"
err_stale_override: "saved to config.yaml, but the stale session override was not cleared ({error})"
next_turn_only: " (next turn only — restores after one response)"
switch_cancelled: "🟡 Model switch cancelled. Current model unchanged ({model})."
guard_confirm: "⚠️ **{title}**\n\n{message}\n\n_Text fallback: reply `{prefix}approve` to switch or `{prefix}cancel` to keep the current model._"
agents:
header: "🤖 **Active Agents & Tasks**"
active_agents: "**Active agents:** {count}"
this_chat: " · this chat"
more: "... and {count} more"
running_processes: "**Running background processes:** {count}"
async_jobs: "**Gateway async jobs:** {count}"
background_delegations: "**Background delegations:** {count}"
none: "No active agents or running tasks."
state_starting: "starting"
state_running: "running"
no_progress: " · no progress {seconds}s"
quiet_for: " · quiet {seconds}s"
between_turns: "between turns"
child_row: " - child {index}: {calls} api calls · {doing}"
active_ago: " · active {seconds}s ago"
approve:
no_pending: "No pending command to approve."
once_singular: "✅ Command approved. The agent is resuming..."
once_plural: "✅ Commands approved ({count} commands). The agent is resuming..."
session_singular: "✅ Command approved (pattern approved for this session). The agent is resuming..."
session_plural: "✅ Commands approved (pattern approved for this session) ({count} commands). The agent is resuming..."
always_singular: "✅ Command approved (pattern approved permanently). The agent is resuming..."
always_plural: "✅ Commands approved (pattern approved permanently) ({count} commands). The agent is resuming..."
background:
usage: "Usage: /bg <prompt>\nExample: /bg Summarize the top HN stories today\n\nRuns the prompt in a separate session. You can keep chatting — the result will appear here when done."
started: "🔄 Background task started: \"{preview}\"\nTask ID: {task_id}\nYou can keep chatting — results will appear when done."
task_finished: "✅ Background task finished"
task_failed: "❌ Background task failed"
exit_code: "exit {code}"
last_output: ". Last output:\n```\n{tail}\n```"
rerun_hint: "\nAsk me to rerun it or show the full log."
no_credentials: "❌ The background task couldn't start because no AI model sign-in is configured. Use /login, or run `hermes setup` on the host."
complete_header: "✅ Background task complete\nPrompt: \"{preview}\"\n\n"
no_response: "(No response generated)"
failed: "❌ Your background task \"{preview}\" failed before finishing. Send /bg again to retry, or /agents to see what is still running."
final_output: "{header}\n\nFinal output:\n```\n{output}\n```"
still_running: "⏳ Background task still running"
recent_output: "{header}\n\nRecent output:\n```\n{output}\n```"
btw:
usage: "Usage: /btw <question>\nExample: /btw which file was that error in?\n\nAnswers a quick side question about this conversation without interrupting it. For an independent background task, use /bg <prompt>."
no_history: "No conversation yet — send your question as a normal message instead."
no_provider: "❌ Cannot answer side question: no provider credentials configured."
started: "💬 Side question: \"{preview}\"\nAnswering from a snapshot of this conversation — the current work continues."
answer: "💬 /btw: \"{preview}\"\n\n{answer}"
failed: "❌ /btw failed: \"{preview}\"\n{error}"
branch:
db_unavailable: "Session database not available."
no_conversation: "No conversation to branch — send a message first."
create_failed: "Failed to create branch: {error}"
switch_failed: "Branch created but failed to switch to it."
branched_one: "⑂ Branched to **{title}** ({count} message copied)\nOriginal: `{parent}`\nBranch: `{new}`\nUse `/resume` to switch back to the original."
branched_many: "⑂ Branched to **{title}** ({count} messages copied)\nOriginal: `{parent}`\nBranch: `{new}`\nUse `/resume` to switch back to the original."
branched_thread_one: "⑂ Branched to **{title}** in a new thread ({count} message copied)\nThread: {thread}\nOriginal: `{parent}` — this chat stays on it\nBranch: `{new}`\nUse `/branch --here` to branch in this chat instead."
branched_thread_many: "⑂ Branched to **{title}** in a new thread ({count} messages copied)\nThread: {thread}\nOriginal: `{parent}` — this chat stays on it\nBranch: `{new}`\nUse `/branch --here` to branch in this chat instead."
thread_fallback: "No thread could be created here, so the branch replaced this chat's session."
commands:
usage: "Usage: `/commands [page]`"
skill_header: "⚡ **Skill Commands**:"
default_desc: "Skill command"
none: "No commands available."
header: "📚 **Commands** ({total} total, page {page}/{total_pages})"
nav_prev: "`/commands {page}` ← prev"
nav_next: "next → `/commands {page}`"
out_of_range: "_(Requested page {requested} was out of range, showing page {page}.)_"
compress:
not_enough: "Not enough conversation to compress (need at least 4 messages)."
no_provider: "No provider configured -- cannot compress."
nothing_to_do: "Nothing to compress yet (the transcript is still all protected context)."
aggressive_unsupported: "--aggressive is not supported; use '/compress here [N]' to keep only recent exchanges, or /undo to drop turns."
focus_line: "Focus: \"{topic}\""
summary_failed: "⚠️ Summary generation failed ({error}). {count} historical message(s) were removed and replaced with a placeholder; earlier context is no longer recoverable. Consider checking your auxiliary.compression model configuration."
aborted: "⚠️ Compression aborted ({error}). No messages were dropped — conversation is unchanged. Run /compress to retry, /reset for a clean session, or check your auxiliary.compression model configuration."
aux_failed: "ℹ️ Configured compression model `{model}` failed ({error}). Recovered using your main model — context is intact — but you may want to check `auxiliary.compression.model` in config.yaml."
turnhold_deferred: "ℹ️ Context compression deferred — summary still streaming. Continuing without compression this turn."
failed: "Compression failed: {error}"
hygiene_timeout: "⚠️ Shortening the conversation history took too long, so I skipped it and kept everything as-is. Run /compress to try again or /new to start fresh."
hygiene_timeout_doctor_hint: " If this keeps happening, run `hermes doctor` on the host."
hygiene_failed: "⚠️ Shortening the conversation history failed, so I kept everything as-is. Run /compress to try again or /new to start fresh. If this keeps happening, run `hermes doctor` on the host."
codex_nothing: "🗜️ Nothing to compact: this session runs on the Codex app-server runtime, whose context lives in a Codex-owned thread that only exists while the agent is active. Send a message first, then /compress — or /reset to start fresh."
codex_compacted: "🗜️ Codex app-server thread compacted (thread/compact). The transcript mirror is unchanged by design — the app-server now carries the compacted context."
codex_incomplete: "⚠️ Codex app-server compaction did not complete — the thread is unchanged. Check the app-server logs, retry /compress, or /reset for a clean session."
debug:
upload_failed: "✗ Failed to upload debug report: {error}"
header: "**Debug report uploaded:**"
auto_delete: "⏱ Pastes will auto-delete in 6 hours."
full_logs_hint: "For full log uploads, use `hermes debug share` from the CLI."
share_hint: "Share these links with the Hermes team for support."
report_label: "Report"
deny:
stale: "❌ Command denied (approval was stale)."
no_pending: "No pending command to deny."
denied_singular: "❌ Command denied."
denied_plural: "❌ Commands denied ({count} commands)."
denied_reason_singular: "❌ Command denied. Reason relayed to the agent: \"{reason}\""
denied_reason_plural: "❌ Commands denied ({count} commands). Reason relayed to the agent: \"{reason}\""
fast:
not_supported: "⚡ /fast is only available for OpenAI models that support Priority Processing."
status: "⚡ Priority Processing\n\nCurrent mode: `{mode}`\n\n_Usage:_ `/fast <normal|fast|auto|cold|ultrafast|status>`"
unknown_arg: "⚠️ Unknown argument: `{arg}`\n\n**Valid options:** normal, fast, auto, cold, ultrafast, status"
saved: "⚡ ✓ Priority Processing: **{label}** (saved to config)\n_(takes effect on next message)_"
session_only: "⚡ ✓ Priority Processing: **{label}** (this session only)"
label_fast: "FAST"
label_normal: "NORMAL"
status_fast: "fast"
status_normal: "normal"
picker_title: "⚡ **Priority Processing**\n\nCurrent mode: `{mode}`\n\nPick an option:"
choice_fast: "fast — Priority Processing on"
choice_normal: "normal — standard processing"
choice_auto: "auto — fast for the first seconds of every turn"
choice_cold: "cold — fast for the first turn of a session only"
choice_ultrafast: "ultrafast — OpenAI Ultrafast (GPT-6 Astra, 6x price)"
ultrafast_not_supported: "⚡ Ultrafast is only available for GPT-6 Astra (current model: `{model}`)."
footer:
status: "📎 Runtime footer: **{state}**\nFields: `{fields}`\nPlatform: `{platform}`"
usage: "Usage: `/footer [on|off|status]`"
saved: "📎 Runtime footer: **{state}**{example}\n_(saved globally — takes effect on next message)_"
example_line: "\nExample: `{preview}`"
state_on: "ON"
state_off: "OFF"
goal:
unavailable: "Goals unavailable on this session."
no_goal_set: "No goal set."
paused: "⏸ Goal paused: {goal}"
no_resume: "No goal to resume."
resumed: "▶ Goal resumed: {goal}\nContinuing now — I'll take the next step right away."
invalid: "Invalid goal: {error}"
set: "⊙ Goal set ({budget}-turn budget): {goal}\nI'll keep working until the goal is done, you pause/clear it, or the budget is exhausted.\nControls: /goal status · /goal pause · /goal resume · /goal clear"
gate_add_admin_only: "⛔ /goal gate add requires an explicitly configured gateway admin (allow_admin_from for DMs, group_allow_admin_from for groups)."
help:
header: "📖 **Hermes Commands**\n"
skill_header: "\n⚡ **Skill Commands** ({count} active):"
more_use_commands: "\n... and {count} more. Use `/commands` for the full paginated list."
insights:
invalid_days: "Invalid --days value: {value}"
error: "Error generating insights: {error}"
kanban:
error_prefix: "⚠ kanban error: {error}"
subscribed_suffix: "(subscribed — you'll be notified when {task_id} completes or blocks)"
truncated_suffix: "… (truncated; use `hermes kanban …` in your terminal for full output)"
no_output: "(no output)"
wake:
completed: "completed"
gave_up: "gave up (retries exhausted)"
crashed: "crashed (worker exited); dispatcher will retry"
timed_out: "timed out; dispatcher will retry"
blocked: "blocked; needs attention"
review_requested: "handed off for review; the implementation is done"
changes_requested: "review requested changes (BLOCK); implementation is not approved"
block_loop_detected: "routed to triage after repeated blocks; orchestration attention needed"
status_default: "status changed"
status_joiner: ", "
message: "[kanban] Task {task_id} {status}.\nTitle: {title}\nAssignee: @{assignee}\nBoard: {board}\n\nCheck the result or decide the next step."
handoff: "Result: {summary}"
review_detail: "Review feedback: {reason}\nInspect the existing card and its current review run; return work to the same implementation task. Do not treat this BLOCK as approval and do not create a duplicate task."
guidance: "This is an automatic task-status notification, not a request to decompose the task again. Inspect the current board before creating follow-up tasks; do not recreate tasks or graphs that already exist."
ping:
completed: "✔ {head} done — {title}{handoff}"
review_requested: "👀 {head} ready for review — {title}{handoff}"
changes_default_reason: "reviewer feedback requires changes"
reviewer_suffix: " — reviewer @{reviewer}"
implementer_suffix: " → implementer @{implementer}"
changes_requested: "🛑 {board_tag}Kanban {task_id} review requested changes/BLOCK: {reason}{provenance}"
triage: "🛑 {head} routed to TRIAGE — {why}{recurrences}{reason}"
triage_decision: "needs a human decision"
triage_attention: "for orchestration attention"
triage_recurrences: " (blocked {value}x for the same cause)"
reason_suffix: ": {value}"
failed_n_times: "it failed {count} times in a row"
kept_failing: "it kept failing"
last_error: " (last: {value})"
gave_up: "⛔ {head} is now blocked: {count}{last}. Fix the cause, then `hermes kanban unblock {task_id}` (or `hermes kanban reassign {task_id}`). Logs: `hermes kanban log {task_id}`."
limit_minutes: "its {minutes}-minute limit"
limit_generic: "its time limit"
timed_out: "⏱ {head} ran past {span} and was stopped; it will be retried automatically."
blocked: "⏸ {head} blocked{reason}"
crashed: "✖ {head} — its worker stopped unexpectedly; it will be retried automatically."
status: "🔄 {head} → {status}"
board_tag: "[{board}] "
assignee_tag: "@{assignee} "
head: "{board_tag}{assignee_tag}Kanban {task_id}"
personality:
none_configured: "No personalities configured in `{path}/config.yaml`"
header: "🎭 **Available Personalities**\n"
none_option: "• `none` — (no personality overlay)"
item: "• `{name}` — {preview}"
usage: "\nUsage: `/personality <name>`"
save_failed: "⚠️ Failed to save personality change: {error}"
cleared: "🎭 Personality cleared — using base agent behavior.\n_(takes effect on next message)_"
set_to: "🎭 Personality set to **{name}**\n_(takes effect on next message)_"
unknown: "Unknown personality: `{name}`\n\nAvailable: {available}"
err_config_write: "config write failed"
profile:
header: "👤 **Profile:** `{profile}`"
home: "📂 **Home:** `{home}`"
plain: "Profile: {profile}\nHome: {home}"
reasoning:
level_default: "medium (default)"
level_disabled: "none (disabled)"
scope_session: "session override"
scope_global: "global config"
status: "🧠 **Reasoning Settings**\n\n**Effort:** `{level}`\n**Scope:** {scope}\n**Display:** {display}\n\n_Usage:_ `/reasoning <none|minimal|low|medium|high|xhigh|max|ultra|reset|show|hide> [--global]`"
display_on: "on ✓"
display_off: "off"
display_set_on: "🧠 ✓ Reasoning display: **ON**\nModel thinking will be shown before each response on **{platform}**."
display_set_off: "🧠 ✓ Reasoning display: **OFF** for **{platform}**"
reset_global_unsupported: "⚠️ `/reasoning reset --global` is not supported. Use `/reasoning <level> --global` to change the global default."
reset_done: "🧠 ✓ Session reasoning override cleared; falling back to global config."
unknown_arg: "⚠️ Unknown argument: `{arg}`\n\n**Valid levels:** none, minimal, low, medium, high, xhigh, max, ultra\n**Display:** show, hide\n**Persist:** add `--global` to save beyond this session"
picker_title: "🧠 **Reasoning Settings**\n\n**Effort:** `{level}`\n**Scope:** {scope}\n**Display:** {display}\n\nPick an option:"
choice_none: "none — disable reasoning"
choice_reset: "reset — clear session override"
choice_show: "show reasoning in replies"
choice_hide: "hide reasoning from replies"
set_global: "🧠 ✓ Reasoning effort set to `{effort}` (saved to config)\n_(takes effect on next message)_"
set_global_save_failed: "🧠 ✓ Reasoning effort set to `{effort}` (session only — config save failed)\n_(takes effect on next message)_"
set_session: "🧠 ✓ Reasoning effort set to `{effort}` (session only — add `--global` to persist)\n_(takes effect on next message)_"
quote_label_discord: "-# 💭 Reasoning"
quote_label_md: "> 💭 **Reasoning:**"
more_lines: "\n_... ({count} more lines)_"
block: "💭 **Reasoning:**\n```\n{reasoning}\n```\n\n{response}"
reload_mcp:
cancelled: "🟡 /reload-mcp cancelled. MCP tools unchanged."
always_followup: "ℹ️ Future `/reload-mcp` calls will run without confirmation. Re-enable via `approvals.mcp_reload_confirm: true` in config.yaml."
confirm_prompt: "⚠️ **Confirm /reload-mcp**\n\nReloading MCP servers rebuilds the tool set for this session and **invalidates the provider prompt cache** — the next message will re-send full input tokens. On long-context or high-reasoning models this can be expensive.\n\nChoose:\n• **Approve Once** — reload now\n• **Always Approve** — reload now and silence this prompt permanently\n• **Cancel** — leave MCP tools unchanged\n\n_Text fallback: reply `/approve`, `/always`, or `/cancel`._"
header: "🔄 **MCP Servers Reloaded**\n"
reconnected: "♻️ Reconnected: {names}"
added: "➕ Added: {names}"
removed: "➖ Removed: {names}"
none_connected: "No MCP servers connected."
tools_available: "\n🔧 {tools} tool(s) available from {servers} server(s)"
failed: "❌ MCP reload failed: {error}"
reload_skills:
header: "🔄 **Skills Reloaded**\n"
no_new: "No new skills detected."
total: "\n📚 {count} skill(s) available"
added_header: "➕ **Added Skills:**"
removed_header: "➖ **Removed Skills:**"
item_with_desc: " - {name}: {desc}"
item_no_desc: " - {name}"
failed: "❌ Skills reload failed: {error}"
reset:
header_default: "✨ Session reset! Starting fresh."
header_new: "✨ New session started!"
header_titled: "✨ New session started: {title}"
title_rejected: "\n⚠️ Title rejected: {error}"
title_error_untitled: "\n⚠️ {error} — session started untitled."
title_empty_untitled: "\n⚠️ Title is empty after cleanup — session started untitled."
tip: "\n✦ Tip: {tip}"
restart:
in_progress: "⏳ Gateway restart already in progress..."
restarting: "♻ Restarting gateway. If you aren't notified within 60 seconds, restart from the console with `hermes gateway restart`."
resume:
db_unavailable: "Session database not available."
parse_error: "⚠️ Could not parse `/resume` arguments: {error}.\nUse quotes around titles with spaces, for example: `/resume \"Project A Plan\"`."
matrix_no_named_sessions: "No named sessions found for this Matrix room.\nUse `/title My Session` to name the current room session, `/resume --all` to list all Matrix sessions, or `/resume --cross-room <session name>` to explicitly cross room boundaries."
matrix_blocked_no_origin: "⚠️ Matrix /resume blocked: this named session has no recorded room origin, so Hermes will not resume it inside the current room by default. Use `/resume --cross-room {name}` if you intentionally want to cross room boundaries."
matrix_blocked_other_room: "⚠️ Matrix /resume blocked: that session belongs to a different Matrix room ({room}). Use `/resume --cross-room {name}` if you intentionally want to resume it here."
matrix_cross_room_success: "⚠️ Cross-room resume: resumed **{title}** inside Matrix room **{room}**.\nFuture messages in this room will use that transcript until `/reset` or another `/resume`.{msg_part}"
blocked_not_owner: "⚠️ /resume blocked: '**{name}**' belongs to a different user or chat. You can only resume sessions from this chat."
no_named_sessions: "No named sessions found.\nUse `/title My Session` to name your current session, then `/resume My Session` to return to it later."
all_requires_admin: "_Note: `--all` (cross-chat listing) requires a configured admin; showing this chat's sessions only._"
list_header: "📋 **Named Sessions**\n"
list_item: "• **{title}**{preview_part}"
list_item_numbered: "{index}. **{title}**{preview_part}"
list_preview_suffix: " — _{preview}_"
list_footer: "\nUsage: `/resume <session name>`"
list_footer_numbered: "\nUsage: `/resume <session name>` or `/resume <number>` (e.g. `/resume 1` for the most recent)"
list_failed: "Could not list sessions: {error}"
out_of_range: "Resume index {index} is out of range.\nUse `/resume` with no arguments to see available sessions."
not_found: "No session found matching '**{name}**'.\nUse `/resume` with no arguments to see available sessions."
already_on: "📌 Already on session **{name}**."
switch_failed: "Failed to switch session."
resumed_one: "↻ Resumed session **{title}** ({count} message). Conversation restored."
resumed_many: "↻ Resumed session **{title}** ({count} messages). Conversation restored."
resumed_no_count: "↻ Resumed session **{title}**. Conversation restored."
msg_count_one: " ({count} message)"
msg_count_other: " ({count} messages)"
retry:
no_previous: "No previous message to retry."
unsafe: "Cannot retry that message safely: {error}"
failed_unchanged: "Retry failed; transcript was not changed."
rollback:
not_enabled: "Checkpoints are not enabled.\nEnable in config.yaml:\n```\ncheckpoints:\n enabled: true\n```"
none_found: "No checkpoints found for {cwd}"
invalid_number: "Invalid checkpoint number. Use 1-{max}."
restored: "✅ Restored to checkpoint {hash}: {reason}\nA pre-rollback snapshot was saved automatically."
kept_user_edits: "↷ Kept your hand-edits: {files}\nUse /rollback <N> --all to restore those too."
kept_oversize: "↷ Kept (too large for checkpoints, no stored copy to revert to): {files}"
failed_deletes: "⚠️ Could not remove (left in place): {files}"
restore_failed: "❌ {error}"
diff:
not_enabled: "Checkpoints are not enabled, so there's no session baseline.\nEnable in config.yaml:\n```\ncheckpoints:\n enabled: true\n```\nPlain /diff still works — it uses git directly."
no_changes: "No changes."
failed: "{error}"
err_default: "Could not generate diff"
more_untracked: "\n... and {count} more"
untracked_label: "**Untracked:**"
truncated_note: "\n... (truncated — {lines} lines total; use /diff --stat for a summary)"
set_home:
save_failed: "Failed to save home channel: {error}"
success: "✅ Home channel set to **{name}** (ID: {chat_id}).\nCron jobs and cross-platform messages will be delivered here."
err_missing_platform: "Missing logical platform"
err_relay_unauthenticated: "Relay does not authenticate this logical home target"
context:
header: "🧠 **Context Window**"
model: "Model: `{model}`"
window: "Window: {total} tokens"
in_use: "In use: {used} / {total} ({pct}%)"
bar: "{bar}"
headroom: "Headroom to limit: {headroom} tokens"
threshold: "Auto-compresses at: {threshold} ({threshold_pct}%) — {to_go} to go"
over_threshold: "⚠️ **Over auto-compression threshold ({threshold}, {threshold_pct}%)**"
compressions: "Compressions this session: {count}"
last_savings: "Last compression freed: {savings}% of context"
totals_header: "Session totals (cumulative across {calls} API calls)"
totals_line: "Input {input} · Output {output} · Reasoning {reasoning}"
total_billed: "Total billed: {total}"
throughput_note: "_Throughput, not context size — each call re-sends the window above._"
estimated: "Estimated context: ~{count} tokens across {messages} messages"
detail_after_first: "_(Full compression and throughput stats available after the first agent response)_"
no_data: "No context data available yet. Send a message to start a session."
category:
system_prompt: "System prompt"
tool_definitions: "Tool definitions"
rules: "Rules"
skills: "Skills"
mcp: "MCP"
subagent_definitions: "Subagent definitions"
memory: "Memory"
conversation: "Conversation"
category_header: "Estimated usage by category"
no_data_yet: " (no data yet — send a message first)"
free_space: "Free space"
category_row: "{glyph} {label} ~{tokens} tokens ~{pct}%"
toolset_row: " {toolset} {count} tools ~{tokens} tokens"
not_available: "n/a"
skill_row: " {name} index ~{index_tokens} SKILL.md {md_tokens} tokens"
and_more: " … and {count} more"
toolsets_title: "Toolsets by schema cost (largest first)"
skills_title: "Skills by cost (index = always-on; SKILL.md = cost when loaded)"
window_line: "Context window: {mark}{used} / {max} tokens ({mark}{pct}%)"
source:
local_estimate: "local estimate"
provider_usage: "provider usage"
provider_usage_plus_estimate: "provider usage + estimated new messages"
source_line: "Source: {source}; category counts are local estimates."
hint_all: "Use /context all for per-skill and per-toolset costs."
status:
header: "📊 **Hermes Gateway Status**"
matrix_scope_header: "**Matrix scope:**"
matrix_scope_room: " room: {room}"
matrix_scope_room_id: " room_id: {room_id}"
matrix_scope_thread: " thread_id: {thread_id}"
matrix_scope_mode: " session_scope: {scope}"
matrix_scope_key: " session_key: {session_key}"
session_id: "**Session ID:** `{session_id}`"
title: "**Title:** {title}"
created: "**Created:** {timestamp}"
last_activity: "**Last Activity:** {timestamp}"
model: "**Model:** `{model}`"
model_provider: "**Model:** `{model}` ({provider})"
free_tier: "Nous · free tier · nous/welcome · /login to sign in"
context: "**Context:** {used} / {total} ({pct}%)"
context_used: "**Context:** ~{used} tokens"
tokens: "**Lifetime tokens billed:** {tokens} _(not your current context size; use `/context`)_"
agent_running: "**Agent Running:** {state}"
state_yes: "Yes ⚡"
state_no: "No"
queued: "**Queued follow-ups:** {count}"
platforms: "**Connected Platforms:** {platforms}"
stop:
stopped_pending: "⚡ Stopped. The agent hadn't started yet — you can continue this session."
stopped: "⚡ Stopped. You can continue this session."
no_active: "No active task to stop."
force_stopped_pending: "⚡ Force-stopped. The agent was still starting — session unlocked."
title:
db_unavailable: "Session database not available."
warn_prefix: "⚠️ {error}"
empty_after_clean: "⚠️ Title is empty after cleanup. Please use printable characters."
set_to: "✏️ Session title set: **{title}**"
not_found: "Session not found in database."
current_with_title: "📌 Session: `{session_id}`\nTitle: **{title}**"
current_no_title: "📌 Session: `{session_id}`\nNo title set. Usage: `/title My Session Name`"
topic:
not_telegram_dm: "The /topic command is only available in Telegram private chats."
no_session_db: "Session database not available."
unauthorized: "You are not authorized to use /topic on this bot."
restore_needs_topic: "To restore a session, first create or open a Telegram topic, then send /topic <session-id> inside that topic. To create a new topic, open All Messages and send any message there."
topics_disabled: "Telegram topics are not enabled for this bot yet.\n\nHow to enable them:\n1. Open @BotFather.\n2. Choose your bot.\n3. Open Bot Settings → Threads Settings.\n4. Turn on Threaded Mode and make sure users are allowed to create new threads.\n\nThen send /topic again."
topics_user_disallowed: "Telegram topics are enabled, but users are not allowed to create topics.\n\nOpen @BotFather → choose your bot → Bot Settings → Threads Settings, then turn off 'Disallow users to create new threads'.\n\nThen send /topic again."
enable_failed: "Failed to enable Telegram topic mode: {error}"
bound_status: "This topic is linked to:\nSession: {label}\nID: {session_id}\n\nUse /new to replace this topic with a fresh session.\nFor parallel work, open All Messages and send a message there to create another topic."
thread_ready: "Telegram multi-session topics are enabled.\n\nThis topic will be used as an independent Hermes session. Use /new to replace this topic's current session. For parallel work, open All Messages and send a message there to create another topic."
untitled_session: "Untitled session"
restore_step_1: "1. Create or open a topic. To create a new one, open All Messages and send any message there."
restore_step_2: "2. Send /topic <session-id> inside that topic."
default_title: "Hermes Chat"
root_lobby: "This main chat is reserved for system commands.\n\nTo start a new Hermes chat, open the All Messages topic at the top of this bot interface and send any message there. Telegram will create a new topic for that message; each topic works as an independent Hermes session."
root_new: "To start a new parallel Hermes chat, open the All Messages topic at the top of this bot interface and send any message there. Telegram will create a new topic for it.\n\nEach topic is an independent Hermes session. Use /new inside an existing topic only if you want to replace that topic's current session."
new_header: "Started a new Hermes session in this topic.\n\nTip: for parallel work, open All Messages and send a message there to create a separate topic instead of using /new here. /new replaces the session attached to the current topic."
help: "/topic — enable multi-session DM mode (one bot, many parallel chats)\n\nUsage:\n /topic Enable topic mode, or show status if already on\n /topic help Show this message\n /topic off Disable topic mode and clear topic bindings\n /topic <id> Inside a topic: restore a previous session by ID\n\nHow it works:\n1. Run /topic once in this DM — Hermes checks BotFather Threads\n Settings are enabled and flips on multi-session mode.\n2. Tap All Messages at the top of the bot and send any message.\n Telegram creates a new topic for that message; each topic is\n an independent Hermes session (fresh history, fresh context).\n3. The root DM becomes a system lobby — send /topic, /status,\n /help, /usage there. Normal prompts go in a topic.\n4. /new inside a topic resets just that topic's session.\n5. /topic <id> inside a topic restores an old session into it."
system_topic_name: "System"
system_topic_intro: "System topic for Hermes commands and status."
setup_image_caption: "BotFather → Bot Settings → Threads Settings"
no_chat_id: "Could not determine chat ID."
mode_not_enabled: "Multi-session topic mode is not currently enabled for this chat."
disable_failed: "Failed to disable topic mode: {error}"
mode_disabled: "Multi-session topic mode is now OFF for this chat.\n\nExisting topics in Telegram aren't removed — they'll just stop being gated as independent sessions. The root DM works as a normal Hermes chat again. Run /topic to re-enable later."
status_enabled: "Telegram multi-session topics are enabled."
status_how_to: "To create a new Hermes chat, open All Messages at the top of this bot interface and send any message there. Telegram will create a new topic for it."
previous_unlinked: "Previous unlinked sessions:"
restore_one: "To restore one:"
restore_example: "Example: Send /topic {session_id} inside a topic."
no_unlinked: "No previous unlinked Telegram sessions found."
restore_later: "To restore a previous session later:"
session_not_found: "Session not found: {session_id}"
not_telegram_session: "That session is not a Telegram session and cannot be restored into this topic."
not_your_session: "That session does not belong to this Telegram user."
already_linked: "That session is already linked to another Telegram topic."
restored: "Session restored: {title}"
restored_last_message: "\n\nLast Hermes message:\n{text}"
undo:
nothing: "Nothing to undo."
removed: "↩️ Undid {turns} turn(s) ({count} message(s)).\nBacked up to: \"{preview}\"\nCopy/edit the text above and send it to re-prompt from here."
invalid_count: "Invalid count \"{arg}\" — use /undo or /undo N."
update:
platform_not_messaging: "✗ /update is only available from messaging platforms. Run `hermes update` from the terminal."
not_git_repo: "✗ Not a git repository — cannot update."
hermes_cmd_not_found: "✗ Could not locate the `hermes` command. Hermes is running, but the update command could not find the executable on PATH or via the current Python interpreter. Try running `hermes update` manually in your terminal."
start_failed: "✗ Failed to start update: {error}"
starting: "☤ Starting Hermes update… I'll stream progress here."
reply_failed: "✗ Failed to send response to update process: {error}"
reply_sent: "✓ Sent `{label}` to the update process."
failed_notice: "❌ Hermes update failed; the previous version is still running. Run `hermes update` on the host to see the full error, or try /update again later."
prompt_default: " (default: {default})"
prompt: "☤ **Update needs your input:**\n\n{prompt}{default_hint}\n\nReply `{prefix}approve` (yes) or `{prefix}deny` (no), or type your answer directly."
finished: "✅ Hermes update finished."
timed_out: "❌ Hermes update timed out after 30 minutes."
finished_success: "✅ Hermes update finished successfully."
last_lines: "{msg}\n\nLast lines:\n```\n{tail}\n```"
managed_action: "update Hermes Agent"
not_applicable: "✗ `hermes update` does not apply to this install ({method}).\nUpdate with: {command}"
usage:
rate_limits: "⏱️ **Rate Limits:** {state}"
header_session: "📊 **Session Token Usage**"
label_model: "Model: `{model}`"
label_input_tokens: "Input tokens: {count}"
label_cache_read: "Cache read tokens: {count}"
label_cache_write: "Cache write tokens: {count}"
label_output_tokens: "Output tokens: {count}"
label_total: "Total: {count}"
label_api_calls: "API calls: {count}"
label_cost: "Cost: {prefix}${amount}"
label_cost_included: "Cost: included"
label_context: "Context: {used} / {total} ({pct}%)"
label_compressions: "Compressions: {count}"
breakdown_header: "🧩 **Context breakdown** _(estimated)_"
breakdown_line: "• {label}: ~{count} ({pct}%)"
breakdown_cat_system_prompt: "System prompt"
breakdown_cat_tool_definitions: "Tool definitions"
breakdown_cat_rules: "Rules"
breakdown_cat_skills: "Skills"
breakdown_cat_mcp: "MCP"
breakdown_cat_subagent_definitions: "Subagent definitions"
breakdown_cat_memory: "Memory"
breakdown_cat_conversation: "Conversation"
header_session_info: "📊 **Session Info**"
label_messages: "Messages: {count}"
label_estimated_context: "Estimated context: ~{count} tokens"
detailed_after_first: "_(Detailed usage available after the first agent response)_"
no_data: "No usage data available for this session."
unknown_subcommand: "Unknown /usage subcommand: `{args}`. Try `/usage` or `/usage reset [--force]`."
reset_wrong_provider: "Banked usage resets are only available on the openai-codex provider. Switch with `/model` first."
credits:
not_logged_in: "Not logged into Nous Portal. Log in to see your credit balance and top up."
verbose:
not_enabled: "The `/verbose` command is not enabled for messaging platforms.\n\nEnable it in `config.yaml`:\n```yaml\ndisplay:\n tool_progress_command: true\n```"
mode_off: "⚙️ Tool progress: **OFF** — no tool activity shown."
mode_new: "⚙️ Tool progress: **NEW** — shown when tool changes (preview length: `display.tool_preview_length`, default 40)."
mode_all: "⚙️ Tool progress: **ALL** — every tool call shown (preview length: `display.tool_preview_length`, default 40)."
mode_verbose: "⚙️ Tool progress: **VERBOSE** — every tool call with full arguments."
mode_log: "⚙️ Tool progress: **LOG** — silent in chat; tool calls written to ~/.hermes/logs/tool_calls.log."
saved_suffix: "_(saved for **{platform}** — takes effect on next message)_"
save_failed: "_(could not save to config: {error})_"
voice:
enabled_voice_only: "Voice mode enabled.\nI'll reply with voice when you send voice messages.\nUse /voice tts to get voice replies for all messages."
disabled_text: "Voice mode disabled. Text-only replies."
tts_enabled: "Auto-TTS enabled.\nAll replies will include a voice message."
status_mode: "Voice mode: {label}"
status_channel: "Voice channel: #{channel}"
status_participants: "Participants: {count}"
status_member: " - {name}{status}"
speaking: " (speaking)"
enabled_short: "Voice mode enabled."
disabled_short: "Voice mode disabled."
label_off: "Off (text only)"
label_voice_only: "On (voice reply to voice messages)"
label_all: "TTS (voice reply to all messages)"
help: "{toggle}\n\n**How /voice works**\n• `/voice on` — voice reply when you send a voice message\n• `/voice tts` — voice reply to *every* message\n• `/voice off` — back to text-only replies\n• `/voice status` — show the current mode\n• `/voice` (no argument) — quick toggle between on and off{channels}"
help_channels: "\n\n**Live voice channels (Discord)**\n• Join a voice channel first, then `/voice channel` — I'll join, listen, and speak my replies\n• `/voice leave` — disconnect from the voice channel"
transcript_echo_short: "🎙️ \"{text}\""
channel_unsupported: "Voice channels are not supported on this platform."
channel_guild_only: "This command only works in a Discord server."
channel_not_in_voice: "You need to be in a voice channel first."
channel_join_failed: "Failed to join voice channel: {error}"
channel_deps_missing: "Voice dependencies are missing (PyNaCl / davey). Run: `{hint}`"
channel_join_permissions: "Failed to join voice channel. Check bot permissions (Connect + Speak)."
channel_joined: "Joined voice channel **{name}**.\nI'll speak my replies and listen to you. Use /voice leave to disconnect."
channel_not_joined: "Not in a voice channel."
channel_left: "Left voice channel."
transcript_echo: "**[Voice]** <@{user}>: {text}"
yolo:
disabled: "⚠️ YOLO mode **OFF** for this session — dangerous commands will require approval."
enabled: "⚡ YOLO mode **ON** for this session — all commands auto-approved. Use with caution."
shared:
session_db_unavailable: "Session database not available."
session_db_unavailable_prefix: "Session database not available"
session_not_found: "Session not found in database."
warn_passthrough: "⚠️ {error}"
unknown_error: "unknown error"
error_prefix: "Error: {error}"
history_unreadable: "⚠️ I can't read this conversation's history right now (your earlier messages exist but cannot be loaded). Run `hermes doctor --fix` on the host, or use /new to start fresh."
agent_running_retry_later: "Agent is running — wait for the turn to finish, then /{command}."
none_marker: "(none)"
none_value: "none"
unknown_value: "unknown"
exec_approval:
header: "Hermes wants to run a command that needs your OK"
reason_label: "Why it was flagged"
default_reason: "dangerous command"
smart_deny_line: "Smart DENY: owner override applies to this one operation only."
smart_deny_heading: "⚠️ **Smart DENY — owner override for one operation:**"
timed_out_notice: "⌛ Approval timed out after {window} — the command was NOT run. Ask me to try again if you still want it, or raise approvals.timeout in config.yaml."
timed_out_card_body: "{notice}\n```\n{command}\n```"
action_once: "Allow Once"
action_session: "Allow Session"
action_always: "Always Allow"
action_deny: "Deny"
deadline: "If you don't answer within {window} it will NOT run."
text_choice_once: "Reply `{prefix}approve` to run it once"
text_choice_session: "`{prefix}approve session` to allow this pattern for the rest of this session"
text_choice_always: "`{prefix}approve always` to allow it permanently"
text_choice_deny: "`{prefix}deny` to cancel"
text_choice_joiner: ", "
text_choice_last: ", or {choice}.\n"
text_body: "{heading}\n```\n{command}\n```\n{reason_label}: {reason}\n\n"
notify:
media_kind:
audio: "audio"
video: "video"
file: "file"
image: "image"
media_delivery_failed: "⚠️ Couldn't deliver the {kind} attachment."
media_delivery_failed_with_name: "⚠️ Couldn't deliver the {kind} attachment ({file_name})."
delivery_failed_retry: "⚠️ Message delivery failed after multiple attempts. Please try again — your request was processed but the response could not be sent."
plain_fallback_prefix: "(Response formatting failed, plain text:)\n\n{content}"
turn_error_no_details: "no details available"
turn_error_detailed: "Sorry, I encountered an error ({error_type}).\n{error_detail}\nTry again or use /reset to start a fresh session."
turn_error: "Sorry, I encountered an error."
context_injection_refused: "Context injection refused."
no_home_channel: "📬 No home channel is set for {platform}. A home channel is where Hermes delivers cron job results and cross-platform messages.\n\nType {sethome_cmd} to make this chat your home channel, or ignore to skip."
unauthorized:
action_notice: "This bot is private and you're not on its allowed list. If you own it, run `hermes pairing approve {platform} <request-id>` on the host (`hermes pairing list` shows the id)."
admin_only_can_run: "You can run: "
admin_only_whoami_hint: ". Use /whoami for the full list."
admin_only_none: "No slash commands are enabled for non-admins on this platform. Ask an admin to add you to allow_admin_from or to set user_allowed_commands."
admin_only: "⛔ /{command} is admin-only here. {suffix}"
owner_hint_env: "add the ID to {env_var} in {home}/.env and restart the gateway"
owner_hint_generic: "add the ID to this platform's allowed-users list and restart the gateway"
owner_notice: "Dropped a message from unrecognized {platform} user {who}. If that is you or someone you trust, {allowlist}; or set `unauthorized_dm_behavior: pair` for {platform} in {home}/config.yaml so unknown senders receive a pairing code you can approve with `hermes pairing approve {platform} <code>`."
picker:
page_info: " ({start}–{end} of {total})"
clarify:
hint_single: "Reply with the number, the option text, or your own answer."
hint_multi: "Multiple selections allowed — reply with the numbers separated by commas or spaces (e.g. \"1, 3\"), the option text, or your own answer."
option_line: " {index}. {choice}"
question: "❓ {question}"
answered: "✅ answered: {response}"
cancelled: "↩️ Clarification cancelled — your message will be handled as a follow-up."
expired: "⏳ This prompt expired — please send a new request."
native_stream_placeholder: "💬 Waiting for your choice..."
skip_hint: 'Reply "skip" to skip this question.'
errors:
rate_limited: "⏱️ The AI model service is rate-limiting requests. Wait a moment, then use /retry."
auth_failed: "⚠️ Sign-in to the AI model service failed. Use /login to sign in again, or ask whoever runs this bot to run `hermes doctor` on the host."
bad_request: "⚠️ The AI model service rejected this request. Try rephrasing your message, or use /model to switch models."
connection_interrupted: "⚠️ The connection to the AI model service was interrupted mid-request — usually transient. Use /retry to try again; if it keeps happening, run `hermes doctor` on the host."
unreachable: "⚠️ The AI model service isn't reachable right now — the configured model endpoint is not running or is unreachable. Wait a moment and use /retry; if it persists, run `hermes doctor` on the host."
connection_unknown: "⚠️ Hermes could not reach the AI model service (no further detail from the SDK). Use /retry to try again; if it persists, run `hermes doctor` on the host."
context_overflow: "⚠️ This conversation has grown too long for me to read all at once. Use /compress to shorten the history, or /new to start a fresh conversation."
usage_limit_resets: "⏱️ The AI model service's usage limit is reached; it resets in {window}. Use /retry after that, or /model to switch models."
provider_kept_failing: "⚠️ The AI model service kept failing. Use /retry to try again, or /model to switch models. Details are in the gateway log (`hermes logs`)."
session_storage_unavailable_disk: "⚠️ Session storage was temporarily unavailable, so this turn was stopped to protect your conversation history. Please check available disk space, then send your message again."
session_storage_unavailable: "⚠️ Session storage was temporarily unavailable, so this turn was stopped to protect your conversation history. Your message should already be saved — please send it again in a moment."
generic_failed: "⚠️ Something went wrong and I couldn't finish this reply. Use /retry to try again, or /new to start a fresh conversation. Technical details are in the gateway log (`hermes logs`)."
interrupted_before_start: "⚠️ Your message was interrupted before processing started (likely by a recent /stop). Please send it again."
processing_incomplete: "processing incomplete"
stopped_before_finishing: "⚠️ I had to stop before finishing{reason}. Use /retry to try again, or /compress if this conversation has grown very long."
no_response: "⚠️ Processing completed but no response was generated. This may be a transient error — try sending your message again."
previous_turn_cleanup: "⚠️ Your message wasn't processed (the previous turn was still being cleaned up). Please send it again."
unexpected_silence: "⚠️ The model returned only a silence marker for a message that needed a reply. Try again or rephrase."
empty_response_model_label: "The model"
hint_auth: " Your sign-in to the AI model service has expired or the API key is wrong. Use /login here, or run `{relogin}` on the host."
hint_quota: " Your AI model service balance or quota is used up. Top it up on the service's website, or use /model to switch models."
hint_overloaded: " The AI model service is temporarily overloaded. Wait a moment, then use /retry."
hint_rate_limited: " You are being rate-limited. Please wait a moment and try again."
hint_usage_limit_resets: " Your plan's usage limit has been reached. It resets in ~{hours}h."
hint_usage_limit: " Your plan's usage limit has been reached. Please wait until it resets."
hint_rejected: " The AI model service rejected the request."
generic_failed_with_hint: "⚠️ Something went wrong and I couldn't finish this reply.{hint}\nUse /retry to try again, or /new to start a fresh conversation. Technical details are in the gateway log (`hermes logs`)."
history_unavailable: "⚠️ This session's history is temporarily unavailable, so this message was not processed. Ask the operator to inspect state.db, then resend after it is healthy. Use /reset only if you intentionally want to start a new conversation."
no_credentials: "⚠️ I couldn't connect to the AI model service, so this message wasn't processed. Use /login to sign in again, or /model to pick a different model. If it keeps failing, run `hermes doctor` on the host."
skills:
disabled: "The **{name}** skill is installed but disabled.\nEnable it with: `hermes skills config`"
not_installed: "The **{name}** skill is available but not installed.\nInstall it with: `hermes skills install {install_name}`"
disabled_for_platform: "The **{name}** skill is disabled for {platform}.\nEnable it with: `hermes skills config`"
stacked_disabled: "The **{names}** skill(s) in this stacked invocation are disabled for {platform}.\nEnable them with: `hermes skills config`"
stacked_load_failed: "Failed to load stacked skills for /{command}."
gate_off: "Skill write approval is off (skills.write_approval). Enable it with /skills approval on, then review staged writes here with /skills pending."
unknown_subcommand: "Unknown /skills subcommand on this platform. Use: pending, approve <id>, reject <id>, diff <id>, approval <on|off>. (Search/install are CLI-only.)"
diff_truncated: "\n… (truncated — full diff in ~/.hermes/pending/skills/{pending_id}.json)"
busy:
action_restarting: "restarting"
action_shutting_down: "shutting down"
reject_model: "Agent is running — wait or /stop first, then switch models."
reject_codex_runtime: "Agent is running — wait or /stop first, then change runtime."
reject_moa: "Agent is running — wait or /stop first, then run /moa."
drain_queued: "⏳ Gateway {action} — queued for the next turn after it comes back."
drain_rejected: "⏳ Gateway is {action} and is not accepting another turn right now."
drain_rejected_new_work: "⏳ Gateway is {action} and is not accepting new work right now."
demoted_tail: " — your message is queued for when it finishes (use /stop to cancel everything)."
elapsed_min: "{minutes} min elapsed"
running_tool: "running: {tool}"
ack_detail_wrap: " ({detail})"
ack_detail_joiner: ", "
ack: "{head}{detail}{tail}"
steered_subagents_head: "⏩ Steered into current run and its active subagent(s)"
steered_subagents_tail: ". Your message arrives after their next tool call."
steered_head: "⏩ Steered into current run"
steered_tail: ". Your message arrives after the next tool call."
redirected_head: "↪ Redirected current run"
redirected_tail: ". I'll adjust using your correction."
subagent_working_head: "⏳ Subagent working"
compressing_head: "⏳ Compressing context"
queued_head: "⏳ Queued for the next turn"
queued_tail: ". I'll respond once the current task finishes."
interrupting_head: "⚡ Interrupting current task"
interrupting_tail: ". I'll respond to your message shortly."
slash_rejected: "⏳ Agent is running — `/{command}` can't run mid-turn. Wait for the current response or `/stop` first."
reject_goal: "Agent is running — use /goal status / pause / clear / wait mid-run, or /stop before setting a new goal."
reject_loop: "Agent is running — use /loop status / pause / stop mid-run, or /stop before setting a new loop."
draining_maintenance: "⏳ This agent is draining for a maintenance action and isn't accepting new turns right now. It'll be back in a moment — please resend shortly."
another_turn_running: "⏳ Another turn is still running on this session. To protect the transcript, this message was not processed. Wait for the active turn to finish, then resend it."
behavior_queue_short: "queues for next turn"
behavior_queue_long: "Messages will be queued for the next turn while Hermes is busy."
behavior_steer_short: "steers into current run (after next tool call)"
behavior_steer_long: "Messages will be steered into the current run (after the next tool call)."
behavior_interrupt_short: "interrupts current run"
behavior_interrupt_long: "Messages will interrupt the current run while Hermes is busy."
status: "**Busy input mode: `{mode}`\nMessages while busy: _{behavior}_\nChange with `/busy queue`, `/busy steer`, or `/busy interrupt`."
unknown_mode: "Unknown mode `{arg}`. Use `/busy queue`, `/busy steer`, or `/busy interrupt`."
save_failed: "Busy input mode could not be saved to config. Mode unchanged."
set: "Busy input mode set to **`{mode}`** (saved).\n_{behavior}_"
pause:
resumed: "▶️ Resumed — new work is accepted again."
not_paused: "Hermes wasn't paused."
reason_suffix: " (reason: {reason})"
already_paused: "⏸️ Hermes is already paused{suffix}. Use `/pause off` to resume."
paused: "⏸️ Paused{suffix}. New cron/kanban/gateway work is on hold; in-flight work finishes normally. Use `/pause off` to resume."
queue:
usage: "Usage: /queue <prompt>"
queued: "Queued for the next turn."
queued_depth: " ({depth} queued)"
steer:
usage: "Usage: /steer <prompt>"
usage_idle: "Usage: /steer <prompt> (no agent is running; sending as a normal message)"
queued_starting: "Agent still starting — /steer queued for the next turn."
queued_no_agent: "No active agent — /steer queued for the next turn."
failed: "⚠️ Steer failed: {error}"
rejected_empty: "Steer rejected (empty payload)."
target_subagents: "run and its active subagent(s)"
target_run: "run"
queued: "⏩ Steer queued into current {target} — arrives after the next tool call: '{preview}'"
suggestions:
failed: "Suggestions command failed: {error}"
blueprint:
failed: "Cron blueprint command failed: {error}"
confirm:
destructive_prompt: "⚠️ **Confirm /{command}**\n\n{detail}\n\nChoose:\n• **Approve Once** — proceed this time only\n• **Always Approve** — proceed and silence this prompt permanently\n• **Cancel** — keep current conversation\n\n_Text fallback: reply `{prefix}approve`, `{prefix}always`, or `{prefix}cancel`._"
destructive_optout: "\n\nℹ️ Future /clear, /new, /reset, and /undo will run without confirmation. Re-enable via `approvals.destructive_slash_confirm: true` in config.yaml."
destructive_optout_save_failed: "\n\n⚠️ Could not save that preference (config.yaml is not writable), so /clear, /new, /reset, and /undo will ask again next time. To silence it permanently, set `approvals.destructive_slash_confirm: false` in config.yaml."
cancelled: "🟡 /{command} cancelled. Conversation unchanged."
detail_new: "This starts a fresh session and discards the current conversation history."
detail_undo_one: "This removes the last user/assistant exchange from history."
detail_undo_many: "This removes the last {count} user turns from history."
hooks:
command_blocked: "Command `/{command}` was blocked by a hook."
prompt_command:
start_failed: "Could not start /{command} — please try again."
learn:
ack_described: "Learning a skill from what you described…"
ack_conversation: "Learning a skill from this conversation…"
plan:
ack_task: "Planning: {task}"
ack_conversation: "Planning from this conversation's context…"
init:
start_failed: "Could not start /init — please try again."
ack_update: "Updating AGENTS.md from a project scan…"
ack_generate: "Generating AGENTS.md from a project scan…"
moa:
prepare_failed: "Failed to prepare MoA turn."
quick_command:
no_output: "Command returned no output."
timed_out: "Quick command timed out (30s)."
error: "Quick command error: {error}"
no_command: "Quick command '/{command}' has no command defined."
unsupported_type: "Quick command '/{command}' has unsupported type (supported: 'exec', 'alias')."
no_target: "Quick command '/{command}' has no target defined."
unknown_command: "Unknown command `/{command}`. Type /commands to see what's available, or resend without the leading slash to send as a regular message."
session:
auto_reset_notice: "◐ Session reset after being stopped. Conversation history cleared.\nUse /resume to browse and restore a previous session.\n"
auto_reset_context_exhausted: "\n\n🔄 Session auto-reset — the conversation exceeded the maximum context size and could not be compressed further. Your next message will start a fresh session."
info_model: "◆ Model: `{model}`"
info_provider: "◆ Provider: {provider}"
info_context: "◆ Context: {tokens} tokens ({source})"
info_acting_model: "◆ Acting model (billed for the run): {provider}:{model}"
info_endpoint: "◆ Endpoint: {url}"
proxy:
requires_aiohttp: "⚠️ Proxy mode requires aiohttp. Run: {hint}"
url_missing: "⚠️ Proxy URL not configured (GATEWAY_PROXY_URL or gateway.proxy_url)"
http_error: "⚠️ Proxy error ({status}): {error}"
closed_early: "⚠️ Proxy connection closed before the response completed"
connection_error: "⚠️ Proxy connection error: {error}"
no_response: "(No response from remote agent)"
progress:
status_fallback_long: "still on it"
status_fallback_short: "one sec"
stuck_warning: "⚠️ I seem to be stuck (no activity for {minutes} min). If nothing happens in the next {remaining} min I'll give up on this task. You can keep waiting, send /stop to cancel it, or /new to start a fresh conversation."
timeout_header: "⏱️ Agent inactive for {minutes} min — no tool calls or API responses."
timeout_stuck_tool: "The agent appears stuck on tool `{tool}` ({seconds}s since last activity, {progress})."
timeout_last_activity: "Last activity: {activity} ({seconds}s ago, {progress}). The agent may have been waiting on an API response."
timeout_hint: "To increase the limit, set agent.gateway_timeout in config.yaml (value in seconds, 0 = no limit) and restart the gateway.\nTry again, or use /reset to start fresh."
working_heartbeat: "⏳ Working — {minutes} min{detail}"
thinking_prefix: "💬 {text}"
tool_head: "{emoji} {tool}"
tool_verbose: "{emoji} {tool}({keys})\n{args}"
tool_preview: "{emoji} {tool}: \"{preview}\""
tool_pending: "{emoji} {tool}..."
task_status_running: "running"
task_status_complete: "complete"
task_status_error: "error"
task_line: "- {title} - {status}"
task_card_title: "Hermes is working"
pairing:
validity_hour: "{hours} hour"
validity_hours: "{hours} hours"
code_reply: "Hi! I don't recognize you yet, so I can't reply until the person running this bot approves you.\n\nYour pairing code: `{code}` (valid for {validity})\n\nIf you run this bot, open a terminal and run: `{approve_cmd}`. Otherwise send that command to the bot owner. After approval, send your message again."
rate_limited: "Too many pairing requests right now. Wait a few minutes, then send your message again."
shutdown:
action_shutting_down: "shutting down"
action_restarting: "restarting"
cron_interrupted: "⚠️ Scheduled job '{job}' was cut short because Hermes is {action}; no result this run. It will run again on schedule, or run it now with `hermes cron run {job}` once Hermes is back."
notice_shutdown: "⚠️ Hermes is shutting down — your current task will be interrupted. When it is back online, send any message and I'll try to pick up where we left off."
notice_restart: "⚠️ Hermes is restarting — your current task will be interrupted. Send any message after the restart and I'll try to resume where you left off."
startup:
handoff_thread_title: "Hermes — {title}"
restarted: "♻ Gateway restarted successfully. Your session continues."
free_tier_line: "Inference: Nous free tier (nous/welcome). Sign in for more: /login"
online: "♻️ Gateway online — Hermes is back and ready."
db_corrupt: "⚠️ Session database corruption detected. Messages may not be persisted. Recovery options:\n1. Run `hermes {profile_arg}doctor --fix`\n2. Stop the gateway, then recover with:\n hermes {profile_arg}sessions recover --source {db_path} --inspect-only\n (if it reports recoverable) hermes {profile_arg}sessions recover --source {db_path} --output recovered-state.db\n — recovery snapshots the damaged file first; do NOT run `sqlite3 ... \".recover\"` against the live state.db, a vulnerable sqlite3 CLI can corrupt it further\n3. Restore from a backup in {backups_dir}/\nRun `hermes {profile_arg}doctor` for sanitized diagnostics."
db_fts_corrupt: "⚠️ Session database reported a corruption error confined to the search index (FTS5); the message tables are not damaged. Messages may not be persisted until it is repaired: run `hermes {profile_arg}doctor --fix`, then restart the gateway. Do not run recovery tools or restore a backup unless `hermes {profile_arg}doctor` confirms damage."
db_then_restart: "{action} Then `hermes {profile_arg}gateway restart`."
db_unavailable: "⚠️ Session database unavailable — messages may not be saved and /resume will be empty. Cause: {cause}. {action}"
whoami:
scope_dm: "DM"
scope_group: "group/channel"
header: "**You** — {platform} ({scope})\nUser ID: `{user_id}`\n"
tier_unrestricted: "Tier: unrestricted (no admin list configured for this scope)\nSlash commands: all available"
tier_admin: "Tier: **admin**\nSlash commands: all available"
tier_user: "Tier: user\nSlash commands you can run: {commands}"
platform:
usage: "Usage: /platform <list|pause|resume> [name]\n /platform list — show platform status\n /platform pause <name> — stop retrying a failing platform\n /platform resume <name> — re-queue a paused platform"
header: "**Gateway platforms**"
connected: "Connected: {platforms}"
reason_default: "paused"
reason_manual: "paused via /platform pause"
item_paused: " · {name} — PAUSED ({reason}). Resume with `/platform resume {name}`."
item_retrying: " · {name} — retrying (attempt {attempts})"
none_failed: "Failed/paused: (none)"
usage_action: "Usage: /platform {action} <name>"
unknown: "Unknown platform: {name}"
not_queued_pause: "{name} is not in the retry queue (it's either connected or not enabled)."
already_paused: "{name} is already paused."
paused: "✓ {name} paused. Resume with `/platform resume {name}` or `hermes gateway restart` to reset."
not_queued_resume: "{name} is not in the retry queue — nothing to resume."
already_retrying: "{name} is already retrying — no resume needed."
resumed: "✓ {name} resumed — retrying on next watcher tick."
# /bundles: ``*_plain`` variants are the surface-independent executor text (hermes_cli/slash_exec.py),
# the others the gateway's markdown rendering.
bundles:
unavailable: "Bundles subsystem unavailable: {error}"
default_desc: "Load {count} skills"
none_plain: "No skill bundles installed.\nCreate one with: hermes bundles create <name> --skill <s1> --skill <s2>\nDirectory: {dir}"
header_plain: "Skill Bundles ({count} installed):"
item_plain: "/{slug} — {desc} ({count} skills)"
invoke_hint_plain: "Invoke a bundle with /<slug> to load all its skills."
none: "No skill bundles installed.\nCreate one on the host with:\n `hermes bundles create <name> --skill <s1> --skill <s2>`\nDirectory: `{dir}`"
header: "**Skill Bundles** ({count} installed):"
item: "• `/{slug}` — {desc} _({count} skills)_"
invoke_hint: "Invoke a bundle with `/<slug>` to load all its skills."
memory:
unknown_subcommand: "Unknown /memory subcommand. Use: pending, approve <id>, reject <id>, approval <on|off>."
approvals:
admin_only: "Only gateway admins can change the persistent approval mode."
heartbeat:
unavailable: "Heartbeats unavailable (no session)."
paused: "⏸ Heartbeat paused: {prompt}"
none_set: "No heartbeat set."
no_resume: "No heartbeat to resume."
resumed: "▶ Heartbeat resumed (every {interval}): {prompt}"
cleared: "✓ Heartbeat cleared."
usage: "Usage: /heartbeat every <interval> <prompt> (e.g. /heartbeat every 10m Check CI)\nAlso: /heartbeat status | pause | resume | clear"
interval_too_small: "Interval too small — minimum is {min_seconds}s."
prompt_required: "Usage: /heartbeat every <interval> <prompt> — the prompt is required."
invalid_prefix: "Invalid heartbeat"
set: "♥ Heartbeat set (every {interval}): {prompt}\nFires as a normal turn whenever this session is idle and the interval has elapsed. Lives while the gateway runs — use `hermes cron` for durable schedules."
refine:
unavailable: "Refine unavailable (no session)."
nothing_yet: "Nothing to refine yet — send a message first."
empty: "Nothing to refine yet — the conversation is empty."
start_failed: "/refine failed to start: {error}"
focus_suffix: " (focus: {focus})"
started: "⚗ Reviewing this conversation in the background{focus_suffix} — any memory/skill updates will be reported when done."
review:
unavailable: "Review unavailable (no session)."
nothing_yet: "Nothing to review yet — send a message first."
start_failed: "/review failed to start: {error}"
subgoal:
no_goal: "No active goal. Set one with /goal <text>."
usage_remove: "Usage: /subgoal remove <n>"
remove_not_int: "/subgoal remove: <n> must be an integer (1-based index)."
removed: "✓ Removed subgoal {index}: {text}"
cleared_one: "✓ Cleared {count} subgoal."
cleared_other: "✓ Cleared {count} subgoals."
none_to_clear: "No subgoals to clear."
added: "✓ Added subgoal {index}: {text}"
loop:
unavailable: "Loops unavailable."
no_session: "Loops unavailable (no active session)."
goal_defers_note: "\nNote: an active /goal is driving this session — loop wakeups defer until the goal finishes, pauses, or parks."
save:
no_messages: "No stored messages found for this session ({session_id})."
no_adapter: "Platform adapter not found to send the document."
caption: "Session export: {filename}"
complete: "Export complete."
failed: "Error exporting session: {error}"
sessions:
usage_search: "Usage: `/sessions search <query>`"
all_requires_admin: "_Note: `all` (cross-chat listing) requires a configured admin; showing this chat's sessions only._"
title_matching: "Sessions matching “{query}”"
title_all: "Sessions"
title_named: "Named Sessions"
codex_runtime:
config_load_failed: "❌ Could not load config: {error}"
topup:
header: "💳 **Nous balance**"
portal_link: "Manage billing on the portal: {url}"
hint: "Top up and manage billing in the browser — your balance updates here after."
login:
free_tier_remaining: "\nSome chats are still on the free tier; use /model in those chats to switch."
platform:
whatsapp:
clarify_other_title: "✏️ Other"
clarify_other_description: "Type your own answer"
clarify_list_button: "Choose"
clarify_list_section: "Options"
approve_button: "✅ Approve"
deny_button: "❌ Deny"
slash_confirm_once: "✅ Approve Once"
slash_confirm_always: "🔒 Always"
slash_confirm_cancel: "❌ Cancel"
clarify_type_answer: "✏️ Type your answer:"
approved: "✅ Approved."
denied: "❌ Denied."
approval_expired: "⌛ Approval expired — command was not run (already timed out or resolved elsewhere)."
yuanbao:
slow_response_notice: "This task is a bit involved — still working on it, please hang on..."
owner_command_denied: "⚠️ {command} is only available to the creator in private chat mode"
shared:
truncated_suffix: "\n...(truncated)"
unknown: "unknown"
approval_expired: "⌛ Approval expired — the command was not run (it already timed out or was resolved elsewhere)."
model_picker_expired: "⏳ This model picker expired — please run /model again."
not_authorized: "⛔ Not authorized."
handoff:
default_name: "session"
media:
image_failed: "⚠️ Couldn't deliver the image attachment."
video_failed: "⚠️ Couldn't deliver the video attachment."
file_failed: "⚠️ Couldn't deliver the file attachment ({name})."
attachment_failed: "⚠️ Couldn't deliver the attachment."
weixin:
voice_as_attachment: "[voice message as attachment]"
signal:
batch_pacing_notice: "(More images coming — pausing ~{wait} for Signal rate limit, batch {batch}/{total}.)"
wait_seconds: "{count}s"
wait_minutes: "{count} min"
api_server:
provider_auth_failed: "Provider authentication failed"
provider_rate_limited: "Provider rate-limited"
provider_error_line: "⚠️ {label}: {error}"
transformed_notice: "\n\n[Response transformed after streaming]\n"
discord:
# Exec-approval card / slash-confirm card. Button labels ≤ 80 chars; embed titles ≤ 256.
approval:
question: "Do you want Hermes to run this command?"
requested_command_label: "Requested command:"
confirm_title: "Confirm"
already_resolved: "This approval has already been resolved~"
expired: "⌛ Approval expired — command was not run (already timed out or resolved elsewhere)"
by_user: "{label} by {user}"
resolved_once: "Approved once"
resolved_session: "Approved for session"
resolved_always: "Approved permanently"
resolved_deny: "Denied"
slash_confirm:
already_resolved: "This prompt has already been resolved~"
approve_once: "Approve Once"
always_approve: "Always Approve"
cancel: "Cancel"
resolved_once: "Approved once"
resolved_always: "Always approved"
resolved_cancel: "Cancelled"
# Clarify / update prompts (embed titles + content headers + button labels + footers).
prompt:
truncated_marker: "... [truncated]"
clarify_title: "Hermes needs your input"
clarify_hint_buttons: "Pick one below, or click ✏️ Other to type a custom answer."
clarify_hint_text: "Reply in this channel with your answer."
default_hint: " (default: {default})"
update_title: "Update Needs Your Input"
expired_footer: "⏱ Prompt expired — no action taken"
already_answered: "Already answered~"
affirm: "Yes"
negate: "No"
other: "✏️ Other (type answer)"
clarify_already_answered: "This prompt has already been answered~"
answered_by: "Answered by {user}: {choice}"
awaiting_typed: "Awaiting typed response from {user}…"
# Model / choice pickers (select placeholders ≤ 150, option label/description ≤ 100).
picker:
title: "⚙ Model Configuration"
select_provider: "Current model: `{model}`\nProvider: {provider}\n\nSelect a provider:"
unknown_model: "unknown"
choice_default_title: "Choose an option"
provider_option: "{provider} ({count} models)"
current: "current"
provider_placeholder: "Choose a provider..."
cancel: "Cancel"
model_placeholder: "Choose a model from {provider}"
back: "◀ Back"
switch_anyway: "Switch anyway"
more_available: "{count} more available — type `/model <name>` directly"
select_model: "Provider: **{provider}**\nSelect a model:{extra}"
already_resolved: "Already resolved~"
switching: "Switching to `{model}`..."
switching_title: "⚙ Switching Model"
switch_error: "Error switching model: {error}"
switched_title: "⚙ Model Switched"
expired_toast: "Model selection expired."
cancelled: "Model selection cancelled."
expired: "⏱ Selection expired — no model change."
choice_placeholder: "Choose an option..."
choice_error: "Error applying selection: {error}"
choice_expired: "⏱ Selection expired — no change made."
# Native slash commands: description / parameter description / choice name ≤ 100 chars (baked
# into Discord at sync time; the sync fingerprint includes the language). Names stay identifiers.
command:
new:
description: "Start a new conversation"
followup: "New conversation started~"
reset:
description: "Reset your Hermes session"
followup: "Session reset~"
model:
description: "Show or change the model"
arg_name: "Model name (e.g. anthropic/claude-sonnet-4). Leave empty to see current."
reasoning:
description: "Show/change reasoning effort, or toggle showing it"
arg_effort: "Pick a level, reset the override, or show/hide reasoning. Leave empty to see current."
choice_none: "none — disable reasoning"
choice_ultra: "ultra — maximum reasoning"
choice_reset: "reset — clear this session's override"
choice_show: "show — reveal reasoning in replies"
choice_hide: "hide — hide reasoning from replies"
personality:
description: "Set a personality"
arg_name: "Personality name. Leave empty to list available."
retry:
description: "Retry your last message"
followup: "Retrying~"
undo:
description: "Remove the last exchange"
status:
description: "Show Hermes session status"
followup: "Status sent~"
stop:
description: "Stop the running Hermes agent"
followup: "Stop requested~"
steer:
description: "Inject a message after the next tool call (no interrupt)"
arg_prompt: "Text to inject into the agent's next tool result"
plan:
description: "Write a markdown implementation plan (no execution)"
arg_task: "What to plan. Leave empty to infer from the conversation."
compress:
description: "Compress conversation context"
title:
description: "Set or show the session title"
arg_name: "Session title. Leave empty to show current."
resume:
arg_name: "Session name to resume. Leave empty to list sessions."
usage:
description: "Show token usage for this session"
help:
description: "Show available commands"
insights:
arg_days: "Number of days to analyze (default: 7)"
reload_skills:
description: "Re-scan ~/.hermes/skills/ for new or removed skills"
voice:
description: "Toggle voice reply mode"
arg_mode: "Voice mode: join, channel, leave, on, tts, off, or status"
choice_join: "join — join your voice channel"
choice_channel: "channel — join your voice channel (alias)"
choice_leave: "leave — leave voice channel"
choice_mode_on: "on — voice reply to voice messages"
choice_tts: "tts — voice reply to all messages"
choice_mode_off: "off — text only"
choice_status: "status — show current mode"
update:
followup: "Update initiated~"
restart:
description: "Gracefully restart the Hermes gateway"
followup: "Restart requested~"
approve:
arg_scope: "Optional: 'all', 'session', 'always', 'all session', 'all always'"
deny:
description: "Deny a pending dangerous command"
arg_scope: "Optional: 'all' to deny all pending commands"
thread:
description: "Create a new thread and start a Hermes session in it"
arg_name: "Thread name"
arg_message: "Optional first message to send to Hermes in the thread"
arg_auto_archive: "Auto-archive in minutes (60, 1440, 4320, 10080)"
error_unknown: "unknown error"
failed: "Failed to create thread: {error}"
created: "Created thread {link}"
error_name_required: "Thread name is required."
error_invalid_archive: "auto_archive_duration must be one of: {allowed}."
error_no_channel: "Could not resolve the current Discord channel."
error_dm_not_supported: "Discord threads can only be created inside server text channels, not DMs."
error_no_parent: "Could not determine a parent text channel for the new thread."
error_both_failed: "Discord rejected direct thread creation and the fallback also failed. Direct error: {direct_error}. Fallback error: {fallback_error}"
queue:
description: "Queue a prompt for the next turn (doesn't interrupt)"
arg_prompt: "The prompt to queue"
followup: "Queued for the next turn."
bg:
arg_prompt: "The prompt to run in the background"
followup: "Background task started~"
btw:
description: "Ask a side question about the current conversation"
arg_question: "The side question to answer without interrupting"
followup: "Side question dispatched~"
skill:
description: "Run a Hermes skill"
arg_name: "Which skill to run"
arg_args: "Optional arguments for the skill"
unknown: "Unknown skill: `{name}`. Start typing for autocomplete suggestions."
auto_args_description: "Arguments: {args_hint}"
auto_description_fallback: "Run /{name}"
limits:
response_truncated: "⚠️ **Response truncated** — this reply exceeded the delivery limit ({max_messages} messages). {dropped_chars} characters were not delivered; the full response is in the session logs."
forum:
default_title: "New Post"
voice:
ack_1: "Let me look into that."
ack_2: "One moment."
ack_3: "Checking on that now."
ack_4: "Give me a sec."
ack_5: "On it."
left_inactivity: "Left voice channel (inactivity timeout)."
security:
unauthorized_slash_alert: "⚠️ Unauthorized Discord slash attempt\nUser: {user_name} ({user_id})\nChannel: {chan_id} (guild {guild_id})\nCommand: {command_text}\nReason: {reason}"
thread:
seed_message: "🧵 Thread created by Hermes: **{name}**"
default_name: "Hermes"
handoff_seed: "🧵 Hermes handoff: **{name}**"
auto_create_failed: "⚠️ Hermes could not create a Discord thread for this message, so the request was not processed. Please retry."
auto_create_failed_generic: "The request was not processed. Please retry."
media:
upload_too_large: "⚠️ Could not attach `{filename}` — {size_mb} MB exceeds Discord's {limit_mb} MB upload limit for this channel. Compress the file or share a link instead."
image_skipped_too_large: "⚠️ Skipped `{filename}` — {size_mb} MB exceeds Discord's {limit_mb} MB upload limit."
telegram:
# Inline-keyboard labels: keep short (2-per-row layout truncates around ~20 chars on mobile).
# Every toast (answerCallbackQuery) is cut at 200 chars by the adapter.
approval:
action_once: "✅ Allow Once"
action_session: "✅ Session"
action_always: "✅ Always"
action_deny: "❌ Deny"
toast_invalid_data: "Invalid approval data."
toast_already_resolved: "This approval has already been resolved."
resolved_once: "✅ Approved once"
resolved_session: "✅ Approved for session"
resolved_always: "✅ Approved permanently"
resolved_deny: "❌ Denied"
resolved_generic: "Resolved"
resolved_by_user: "{label} by {user}"
expired: "⌛ Approval expired"
expired_detail: "{label} — no command was waiting. It already timed out (and was denied) or was resolved elsewhere."
slash_confirm:
approve_once: "✅ Approve Once"
always_approve: "🔒 Always Approve"
cancel: "❌ Cancel"
already_resolved: "This prompt has already been resolved."
resolved_once: "✅ Approved once"
resolved_always: "🔒 Always approve"
resolved_cancel: "❌ Cancelled"
# Model / choice pickers (MarkdownV2 bodies; the adapter adds the markup around these).
picker:
expired_rerun: "Picker expired — run the command again."
invalid_selection: "Invalid selection."
expired: "Picker expired."
apply_error: "Error applying selection: {error}"
prev: "◀ Prev"
next: "Next ▶"
back: "◀ Back"
cancel: "✗ Cancel"
more_available: "{count} more available — type `/model <name>` directly"
title: "Model Configuration"
provider_label: "Provider: {provider}"
select_model: "Select a model:"
current_model: "Current model: {model}"
select_provider: "Select a provider:"
unknown_model: "unknown"
provider_family: "Provider family: {family}"
invalid_model_index: "Invalid model index."
switch_error: "Error switching model: {error}"
switch_failed: "Switch failed."
switched: "Model switched!"
invalid_page: "Invalid page."
expired_model: "Picker expired — use /model again."
provider_not_found: "Provider not found."
switch_anyway: "Switch anyway"
confirm_selection: "Confirm model selection"
group_not_found: "Group not found."
cancelled: "Model selection cancelled."
# Update (/update watcher) and clarify prompts.
prompt:
default_hint: " (default: {default})"
update_header: "Update needs your input:"
affirm: "✓ Yes"
negate: "✗ No"
other: "✏️ Other (type answer)"
expired_toast: "⚠️ This prompt expired — please /retry."
expired_body: "⚠️ This question expired or the session reset — please /retry."
type_answer: "✏️ Type your answer in the chat."
awaiting_typed: "Awaiting typed response from {user}…"
invalid_choice: "Invalid choice."
update_sent: "Sent '{answer}' to the update process."
update_answered: "Update prompt answered: {answer}"
affirm_word: "Yes"
negate_word: "No"
media:
download_failed: "⚠️ Couldn't download your {kind}{name} ({error}). Please try sending it again."
named_suffix: " ({name})"
kind_attachment: "attachment"
kind_photo: "photo"
kind_voice: "voice message"
kind_audio: "audio file"
kind_video: "video file"
gmail_triage:
done_send: "✓ sent draft"
done_archive: "✓ archived"
done_draft: "✓ drafted reply"
done_spam: "✓ marked spam"
done_mute: "✓ muted"
done_mute_domain: "✓ muted domain"
done_trust: "✓ trusted"
done_trust_domain: "✓ trusted domain"
done_vip: "✓ marked VIP"
done_vip_domain: "✓ marked VIP domain"
invalid_data: "Invalid gmail-triage data."
unknown_verb: "Unknown verb: {verb}"
script_missing: "❌ {script} missing"
exit_code: "exit {code}"
failed: "❌ {verb} failed: {detail}"
timed_out: "❌ {verb} timed out"
error: "❌ {verb} error: {error}"
appended_by: "— {label} by {user}"
user_fallback: "User"
install_hint: "Run `hermes setup` to install Telegram support."
# Slash-command DESCRIPTIONS (names are identifiers and never translated). Owned by lane W1a; the
# keys below are the ones the Discord native command table shares verbatim with the CLI registry.
slack:
approval:
fallback_text: "⚠️ Command approval required: {command}"
resolved_once: "✅ Approved once by {user}"
resolved_session: "✅ Approved for session by {user}"
resolved_always: "✅ Approved permanently by {user}"
resolved_deny: "❌ Denied by {user}"
confirm:
default_title: "Confirm"
approve_once: "Approve Once"
always_approve: "Always Approve"
cancel: "Cancel"
resolved_always: "🔒 Always approved by {user}"
resolved_cancel: "❌ Cancelled by {user}"
picker:
more_available: "\n*{count} more available — type `/model <name>` directly*"
provider_header: "*⚙ Model Configuration*\nCurrent model: `{model}`\nProvider: {provider}\n\nSelect a provider:{extra}"
model_header: "*⚙ Model Configuration*\n\nProvider: *{provider}*\nSelect a model:{extra}"
choose_provider: "Choose a provider…"
choose_model: "Choose a model from {provider}…"
cancel: "Cancel"
back: "◀ Back"
fallback_provider: "⚙ Model Configuration — select a provider"
fallback_model: "⚙ Model Configuration — {provider}"
cancelled: "❌ Model selection cancelled."
no_models_for_provider: "No models available for `{provider}`."
switching: "⚙ Switching to `{model}`…"
switch_failed: "❌ Model switch failed: {error}"
switch_failed_header: "⚙ Model Switch Failed"
switched_header: "⚙ Model Switched"
clarify:
option_n: "Option {n}"
choice_n: "choice {n}"
other: "✏️ Other…"
expired: "⏳ This prompt expired — please send a new request. (by {user})"
awaiting: "✏️ Awaiting typed answer from {user}…"
resolved: "✅ {user}: {choice}"
feedback:
good: "Good Response"
good_a11y: "Submit positive feedback on this response"
bad: "Bad Response"
bad_a11y: "Submit negative feedback on this response"
status:
thinking: "is thinking..."
still_working: "still working… ({elapsed})"
task_card:
title: "Hermes is working"
slash:
reply_truncated: "\n\n_[Reply truncated: {count} more part(s) exceeded Slack's ephemeral reply limit.]_"
running: "Running `/{command}`…"
handoff:
seed: ":thread: Hermes handoff — *{name}*"
matrix:
handoff:
default_name: "Hermes session"
media:
image_download_failed: "I couldn't download and upload the image to Matrix. The source URL was not shown because it may contain private tokens."
approval:
typed_hint_once: "Reply `!approve` to execute once, or `!deny` to cancel."
typed_hint_session: "Reply `!approve session` to approve this pattern for the session, `!approve` to execute once, or `!deny` to cancel."
typed_hint_always: "Reply `!approve session` to approve this pattern for the session, `!approve always` to approve permanently, `!approve` to execute once, or `!deny` to cancel."
legend_intro: "You can also click the reaction to approve:"
legend_once: "✅ = approve once"
legend_session: "🌀 = approve for this session"
legend_always: "♾️ = approve always"
legend_deny: "❎ = deny"
invalid_reaction: "That reaction is not valid for this approval prompt."
expired: "This approval prompt has expired. Run the command again if you still want to approve it."
picker:
no_models: "No authenticated models are available for this session."
title: "⚙ **Model Configuration**"
current_model: "Current model: `{model}`"
provider: "Provider: {provider}"
react_model: "React to choose a model:"
react_choice: "React to choose."
current_suffix: "{label} ← current"
invalid_model_reaction: "That reaction is not one of the available model choices."
invalid_choice_reaction: "That reaction is not one of the available choices."
verb_switch_model: "switch model"
verb_apply_selection: "apply selection"
failed: "Failed to {action}: {error}"
reaction:
unauthorized: "Only an authorized Matrix user can use these controls."
not_requester: "Only the user who requested this action can use these controls."
feishu:
approval:
# Feishu keeps shorter button words than the shared card (emoji prefixes are added in code).
action_session: "Session"
action_always: "Always"
resolved_once: "Approved once"
resolved_session: "Approved for session"
resolved_always: "Approved permanently"
resolved_deny: "Denied"
resolved_fallback: "Resolved"
resolved_by_user: "{icon} **{label}** by {user}"
update_prompt:
title: "☤ Update Needs Your Input"
default_hint: "\n\nDefault: `{default}`"
answer_yes: "✓ Yes"
answer_no: "✗ No"
word_yes: "Yes"
word_no: "No"
answered_title: "{icon} Update prompt answered: {answer}"
answered_by: "Answered by **{user}**"
media:
gif_downgraded: "[GIF downgraded to file]"
gif_downgraded_caption: "[GIF downgraded to file]\n{caption}"
google_chat:
clarify:
other_button: "Other / type answer"
header: "Question"
question: "❓ {question}"
typing:
thinking: "Hermes is thinking…"
interrupted: "(interrupted)"
no_reply: "(no reply)"
attachment_fallback:
# Shipped in Spanish before i18n; es.yaml keeps that original text.
header: "⚠️ I couldn't attach **{filename}**."
explain: "Google Chat only allows file attachments when the bot has your explicit permission (user OAuth). It's a one-time consent done from this chat."
activate: "**To enable it:** send `/setup-files` and follow the instructions."
host_path: "Meanwhile the file is on the host: `{path}`"
setup_files:
not_configured: "🔧 Native attachment delivery is **not configured**.\n**Step 1 (one-time, on the host):** create OAuth client credentials at https://console.cloud.google.com/apis/credentials → *Create credentials* → *OAuth client ID* → *Desktop app*. Download the JSON. Then on the host run:\n```\npython -m plugins.platforms.google_chat.oauth --client-secret /path/to/client_secret.json\n```\n**Step 2:** come back here and send `/setup-files start`."
start_instructions: "1. Open this URL in your browser and authorize:\n{auth_url}\n\n2. After clicking *Allow*, your browser will fail to load `http://localhost:1/?...&code=...`. That's expected.\n\n3. Copy the entire failed URL from the browser's URL bar and paste it back here as: `/setup-files <PASTE_URL>` (or just the `code=...` value).\n\nTip: the URL contains your access grant — keep it private."
start_failed: "❌ Couldn't generate the OAuth URL. Check the gateway logs and verify the client_secret.json is valid."
exchange_failed: "❌ Token exchange failed. The code may have expired or the URL is malformed. Send `/setup-files start` to get a fresh OAuth URL."
revoke_completed: "Revoke completed (some steps may have been skipped)."
revoked: "Revoked."
helper_error: "❌ Error: {error}"
revoke_error: "❌ Error revoking: {error}"
who_shared: "shared (legacy)"
active: "✅ Native attachment delivery is **active** for `{who}`.\nToken: `{token_path}`\nSend `/setup-files revoke` to disable."
not_authorized_yet: "🔧 Client credentials are stored but you haven't authorized yet. Send `/setup-files start` to begin."
no_client_credentials: "⚠️ No client credentials stored for this profile. Send `/setup-files` (no args) for setup instructions."
done_output: "✅ Done.\n```\n{output}\n```"
authorized: "✅ Authorized! Native attachment delivery is now active. Try asking me to send you a PDF."
exchanged_not_loaded: "⚠️ Token exchanged but the gateway couldn't load the new credentials in-memory. Restart the gateway and the token at `{token_path}` will be picked up.\nHelper output:\n```\n{output}\n```"
teams:
approval:
resolved_once: "✅ Allowed (once)"
resolved_session: "✅ Allowed (session)"
resolved_always: "✅ Always allowed"
resolved_deny: "❌ Denied"
unknown_action: "Unknown action."
requires_allowlist: "⛔ Approval buttons require TEAMS_ALLOWED_USERS to be configured."
summary:
title: "Meeting {meeting_id}"
title_fallback: "summary"
heading: "Summary"
summary_line: "Summary: {summary}"
empty: "No summary available."
none_item: "None"
key_decisions: "Key decisions"
action_items: "Action items"
risks: "Risks"
line:
pending:
still_thinking: "🤔 Still thinking. Tap below to fetch the answer when it's ready."
button_label: "Get answer"
delivered: "Already replied ✅"
interrupted: "Run was interrupted before completion."
expired: "That request has expired — send your message again."
standalone:
attachments_not_deliverable: "[{count} attachment(s) generated; not deliverable from cron]"
wecom:
media:
# Shipped in Chinese before i18n; zh.yaml keeps that original text.
kind_image: "Image"
kind_video: "Video"
kind_voice: "Voice message"
file_too_large: "File size {size_mb}MB exceeds WeCom's 20MB limit and cannot be sent. Try compressing the file or reducing its size."
voice_format_downgraded: "Voice format {content_type} is not supported (WeCom only accepts AMR); sent as a file instead."
size_downgraded: "{label} size {size_mb}MB exceeds the {cap} limit; sent as a file instead."
email:
image_line: "Image: {url}"
standalone_subject: "Hermes Agent"
dingtalk:
emotion:
thinking: "🤔Thinking"
done: "🥳Done"
irc:
quit_message: "Hermes Agent shutting down"
standalone_quit: "delivered"
homeassistant:
notification_title: "Hermes Agent"
relay:
confirm_approve_once: "Approve Once"
confirm_always_approve: "Always Approve"
confirm_cancel: "Cancel"
prompt_other: "✏️ Other (type answer)"
prompt_expired: "⌛ That prompt is no longer waiting for an answer. Send your reply as a normal message."
slash:
approve:
description: "Approve a pending dangerous command"
bg:
description: "Run a prompt in a separate background session"
insights:
description: "Show usage insights and analytics"
reload_mcp:
description: "Reload MCP servers from config"
resume:
description: "Resume a previously-named session"
sethome:
description: "Set this chat as the home channel"
update:
description: "Update Hermes Agent to the latest version"
category:
session: "Session"
configuration: "Configuration"
info: "Info"
tools_skills: "Tools & Skills"
plugins: "Plugins"
exit: "Exit"
context: "Context"
background_automation: "Background & Automation"
shared:
usage_suffix: "{description} (usage: /{name} {args_hint})"
alias_for: "{description} (alias for /{name})"
alias_note: " (alias: {aliases})"
slack_alias_for: "Alias for /{name} — {description}"
plugin_default_desc: "Run /{name}"
plugin_command_desc: "Plugin command"
hermes:
description: "Talk to Hermes or run a subcommand"
start:
description: "Acknowledge platform start pings without a reply"
new:
description: "Start a new session (fresh session ID + history)"
topic:
description: "Enable or inspect Telegram DM topic sessions"
clear:
description: "Clear screen and start a new session"
redraw:
description: "Force a full UI repaint (recovers from terminal drift)"
history:
description: "Show conversation history"
save:
description: "Export the current conversation (bare /save shows usage)"
retry:
description: "Retry the last message (resend to agent)"
prompt:
description: "Compose your next prompt in $EDITOR (markdown), then send it"
undo:
description: "Back up N user turns and re-prompt (default 1)"
title:
description: "Set a title for the current session"
handoff:
description: "Hand off this session to a messaging platform (Telegram, Discord, etc.)"
branch:
description: "Branch the current session (new thread on Discord/Telegram/Slack/Matrix; --here stays here)"
worktree:
description: "Show, list, create, or prune isolated git worktrees"
compress:
description: "Compress conversation context (add 'here [N]' to keep recent N turns; --preview shows what would happen)"
rollback:
description: "List or restore filesystem checkpoints (restores keep your hand-edits; --all overrides)"
snapshot:
description: "Create or restore state snapshots of Hermes config/state"
export:
description: "Export a profile (config, skills, theme) to a shareable archive"
import:
description: "Import a shared profile archive as a new profile"
stop:
description: "Kill all running background processes"
pause:
description: "Pause new work globally (emergency stop); '/pause off' resumes"
deny:
description: "Deny a pending dangerous command (optionally with a reason)"
btw:
description: "Ask a side question about the current conversation without interrupting it"
agents:
description: "Show active agents and running tasks"
journey:
description: "Open the learning journey timeline"
queue:
description: "Queue a prompt for the next turn, or list/edit/rm/move/clear queued prompts"
steer:
description: "Inject a message after the next tool call without interrupting"
goal:
description: "Set a standing goal Hermes works on across turns until achieved"
heartbeat:
description: "Set a recurring prompt that re-enters this session when idle"
refine:
description: "Review this conversation now and save lessons to memory/skills"
review:
description: "Spawn an independent subagent to review the work just discussed (PR, code, docs)"
loop:
description: "Re-run a prompt on a recurring interval in this session"
plan:
description: "Write a markdown implementation plan to .hermes/plans/ without executing anything"
moa:
description: "Run one prompt through the default Mixture of Agents preset, then restore your model"
subgoal:
description: "Add or manage extra criteria on the active goal"
status:
description: "Show session, model, token, and context info"
egress:
description: "Show Docker egress proxy status"
context:
description: "Show detailed context window view with usage gauge, category breakdown, compression stats, and throughput"
whoami:
description: "Show your slash command access (admin / user)"
profile:
description: "Show active profile name and home directory"
sessions:
description: "Browse and resume previous sessions"
config:
description: "Show current configuration"
model:
description: "Switch model (session-scoped; --global to persist)"
codex-runtime:
description: "Toggle codex app-server runtime for OpenAI/Codex models"
personality:
description: "Set a predefined personality"
statusbar:
description: "Toggle the context/model status bar"
battery:
description: "Toggle a color-coded battery indicator in the status bar"
timestamps:
description: "Toggle [HH:MM] timestamps on messages and /history"
diff:
description: "Show git changes in the working directory"
verbose:
description: "Cycle tool progress display: off -> new -> all -> verbose"
focus:
description: "Toggle focus view — show only your prompt and the final response"
footer:
description: "Toggle gateway runtime-metadata footer on final replies"
yolo:
description: "Toggle YOLO mode (skip all dangerous command approvals)"
approvals:
description: "Show or set the persistent dangerous-command approval mode"
reasoning:
description: "Manage reasoning effort and display"
fast:
description: "Fast mode — OpenAI Priority Processing / Anthropic Fast Mode (normal/fast/auto/cold/ultrafast)"
skin:
description: "Show or change the display skin/theme"
indicator:
description: "Pick the TUI busy-indicator style"
voice:
description: "Toggle voice mode"
wake:
description: "Toggle the 'Hey Hermes' wake word listener"
busy:
description: "Control how messages behave while Hermes is working"
tools:
description: "Manage tools: /tools [list|disable|enable] [name...]"
toolsets:
description: "List available toolsets"
skills:
description: "Search, install, inspect, or manage skills"
memory:
description: "Review pending memory writes / toggle the approval gate"
bundles:
description: "List skill bundles (aliases /<name> for multiple skills)"
pet:
description: "Toggle or adopt a petdex mascot (/pet, /pet list, /pet <slug>)"
hatch:
description: "Generate a new petdex pet from a description"
learn:
description: "Learn a reusable skill from anything you describe (dirs, URLs, this chat, notes)"
init:
description: "Generate or update AGENTS.md project instructions from a repo scan"
cron:
description: "Manage scheduled tasks"
suggestions:
description: "Review suggested automations (accept/dismiss)"
blueprint:
description: "Set up an automation from a blueprint template"
curator:
description: "Background skill maintenance (status, run, pin, archive, list-archived)"
kanban:
description: "Multi-profile collaboration board (tasks, links, comments)"
reload:
description: "Reload .env variables into the running session"
reload-mcp:
description: "Reload MCP servers from config"
reload-skills:
description: "Re-scan ~/.hermes/skills/ for newly installed or removed skills"
browser:
description: "Connect browser tools to your live Chromium-family browser via CDP, or switch to Browser Use mode"
plugins:
description: "List installed plugins and their status"
commands:
description: "Browse all commands and skills (paginated)"
help:
description: "Show available commands (/help skills lists skill commands, /help <text> filters)"
palette:
description: "Open the fuzzy command palette (also Ctrl+P)"
restart:
description: "Gracefully restart the gateway after draining active runs"
usage:
description: "Show token usage and rate limits; `reset` redeems a banked Codex limit reset"
subscription:
description: "View your Nous plan and change it in the browser"
login:
description: "Sign in with a Nous account (keeps your connectors)"
topup:
description: "Show your Nous balance and manage billing on the portal"
platforms:
description: "Show gateway/messaging platform status"
platform:
description: "Pause, resume, or list a failing gateway platform"
copy:
description: "Copy the last assistant response to clipboard"
paste:
description: "Attach clipboard image from your clipboard"
image:
description: "Attach a local image file for your next prompt"
version:
description: "Show Hermes Agent version"
debug:
description: "Upload debug report (system info + logs) and get shareable links"
quit:
description: "Exit the CLI (use --delete to also remove session history)"
cli:
billing:
# shared helpers (/topup + /subscription)
load_failed: "{label}: {error}"
load_failed_label: "Couldn't load billing"
not_logged_in: "Not logged into Nous Portal."
run_portal_then: "Run `hermes portal` to log in, then {cmd}."
org_line: "Org: {org}"
org_line_role: "Org: {org} · {role}"
bar_plan_label: "plan"
bar_topup_label: "top-up"
bar_pct_used: " · {pct}% used"
bar_plan: "{label}[{bar}] ${remaining} left of ${total}{pct}"
bar_topup: "{label}[{bar}] ${remaining} · never expires"
total_spendable: "Total spendable: ${amount}"
one_time_setup: "! One-time setup"
allow_remote_spending_title: "Allow Remote Spending"
allow_remote_spending_detail: "Opens your browser to authorize this terminal."
choice_allow_remote_spending: "Allow Remote Spending"
desc_allow_remote_spending: "open your browser to authorize"
choice_not_now: "Not now"
desc_not_now: "cancel"
opening_browser_remote_spending: "Opening your browser to allow Remote Spending…"
couldnt_allow_remote_spending: "Couldn't allow Remote Spending: {error}"
remote_spending_allowed: "Remote Spending allowed."
portal_line: "Portal: {url}"
open_this_url: "Open this URL: {url}"
manage_on_portal_url: "Manage on portal: {url}"
no_manage_url: "No manage URL available — is your portal configured?"
stepup_stale: "Remote Spending still isn't active for this terminal — the authorization didn't take. Retry, or make this change on the portal."
something_went_wrong: "Something went wrong."
under_five: "under $5"
plan_free: "Free"
# /usage balance block
renews_suffix: " · renews {date}"
plan_line: "Plan: {plan}{renews}"
free_models_only_hint: "> Free · free models only. Run /subscription to reach paid models."
low_balance_usage: "! Low balance · {amount} left. Run /topup or /subscription."
usage_cta: "Run /subscription to change plan · /topup to add to your balance"
# shared modal rows
choice_cancel: "Cancel"
choice_back: "Back"
choice_go_back: "Go back"
choice_close: "Close"
desc_do_nothing: "do nothing"
desc_do_not_charge: "do not charge"
choice_manage_on_portal: "Manage on portal"
desc_open_billing_page: "open the billing page in your browser"
# /topup overview (screen 1)
topup_header: "Top up · balance {balance}"
topup_title: "Top up your balance"
auto_reload_on_line: "Auto-reload: on — below {threshold} → reload to {reload_to}"
auto_reload_off_line: "Auto-reload: off"
card_line: "Card: {card}"
no_saved_card_add_funds_hint: "No saved card on file — “Add funds” walks you through adding one."
actions_require_admin: "Billing actions require an org admin/owner."
remote_spending_off_for_org: "Remote spending is off for this org."
killswitch_reason_overview: "A billing admin can turn it on from the portal's Hermes Agent page to add funds here."
killswitch_reason_buy: "A billing admin can turn it on from the portal's Hermes Agent page before adding funds."
add_funds_now_hint: "Add funds now — a single charge, added to your balance today."
refill_when_low_amounts: "Refill when low — charges {reload_to} automatically when your balance falls below {threshold}."
refill_when_low_generic: "Refill when low — charges your card automatically when your balance falls below the amount you set."
choice_add_funds: "Add funds"
desc_add_funds: "a single charge, added to your balance today"
choice_auto_reload: "Auto-reload"
desc_auto_reload: "refill automatically when your balance runs low"
choice_monthly_limit: "Monthly limit"
desc_monthly_limit: "show the monthly spend cap (read-only)"
cancelled: "Cancelled."
cancelled_yellow: "🟡 Cancelled."
cancelled_no_funds: "Cancelled. No funds added."
no_portal_url: "No portal URL available."
complete_in_browser: "Complete billing changes in the browser."
# add-card flow
add_card_first: "Add a card first"
add_card_title: "Add a card"
no_saved_card: "No saved card on file."
no_saved_card_short: "No saved card on file"
no_saved_card_lower: "no saved card"
add_card_once_hint: "Add a card once on the portal billing page — after that you can top up right from the terminal."
add_card_then_recheck: "Add the card on the billing page, then pick “check again” here."
card_found_continuing: "Card found: {card} — continuing."
still_no_card: "Still no card on file — finish adding it on the portal, then check again."
choice_add_card_on_portal: "Add a card on the portal"
desc_add_card_on_portal: "opens the billing page in your browser"
choice_recheck_card: "I've added it — check again"
desc_recheck_card: "re-check for the card and continue"
# buy flow (screens 2–3)
add_funds: "Add funds"
presets: "Presets: {presets}"
run_interactive_to_purchase: "Run this in the interactive CLI to complete a purchase."
desc_one_time_purchase: "one-time credit purchase"
choice_custom_amount: "Custom amount…"
desc_custom_amount: "enter your own amount"
payment_line: "Payment: {card}"
amount_prompt: "Amount (USD):"
invalid_selection: "🔴 Invalid selection."
confirm_purchase: "Confirm purchase"
total_line: "Total: {amount}"
portal_card_will_be_charged: "Your card saved on the portal will be charged."
consent_charge: "By confirming, you allow Nous Research to charge your card."
choice_pay_now: "Pay {amount} now"
desc_submit_charge: "submit the charge"
desc_manage_card_billing: "manage your card / billing in the browser"
run_interactive_to_confirm: "Run in the interactive CLI to confirm a purchase."
pay_question: "Pay {amount}?"
no_charge_id: "No charge id returned; please check the portal."
charge_submitted: "Charge submitted — confirming settlement…"
could_not_check_charge: "🔴 Could not check the charge: {error}"
added_to_balance: "✓ {amount} added to your balance."
still_processing: "🟡 Still processing after 5 minutes — this is a timeout, not a failure. Check /billing or the portal shortly."
# charge outcomes
charge_failed_authentication_required: "🔴 Your bank requires verification (3DS). Complete it on the portal to finish this purchase."
charge_failed_payment_method_expired: "🔴 Your card has expired. Update it on the portal."
charge_failed_card_declined: "🔴 Your card was declined. Try another card on the portal."
charge_failed_generic: "🔴 The charge didn't go through ({reason})."
charge_error_no_payment_method: "💳 No card on file — top up and manage billing on the portal."
charge_error_remote_spending_disabled: "Remote spending is off for this account — a billing admin can turn it on from the portal's Hermes Agent page."
charge_error_role_required: "Adding funds needs an org admin/owner. Ask an admin, or manage on the portal."
charge_error_idempotency_conflict: "🔴 That charge key was already used for a different amount. Start a fresh top-up."
revoked_by_admin: "An admin stopped this terminal's spending."
revoked_by_you: "You stopped this terminal's spending."
revoked_reconnect: "🔴 {who} Reconnect to restore — run `hermes portal` to re-authorize."
session_logged_out: "🔴 Your session was logged out. Run `hermes portal` to log in again."
monthly_cap_reached_headroom: "🔴 Monthly spend cap reached — ${remaining} headroom left."
monthly_cap_reached: "🔴 Monthly spend cap reached."
try_again_in_min: " (try again in ~{minutes} min)"
too_many_charges: "🟡 Too many charges right now{mins}. This isn't a payment failure."
remote_spending_needs_approval: "🔴 Remote Spending needs approval — run /topup to allow it, then retry."
# step-up (Remote Spending) during a charge
your_topup: "your top-up"
charge_stepup_explain: "To charge from this terminal, allow Remote Spending once. It opens your browser to authorize, then {amount} picks up right here."
charge_stepup_noninteractive: "Run `hermes portal` and allow Remote Spending, then retry."
charge_stepup_declined: "No charge made. Run /topup when you want to allow Remote Spending."
charge_stepup_not_granted: "Couldn't allow Remote Spending — an org admin or owner has to approve it. Your card was not charged."
allowed_but_org_off: "Remote Spending is allowed for this terminal, but it's still off for this org. A billing admin can turn it on from the portal's Hermes Agent page, then run /topup again."
allowed_no_card: "Remote Spending allowed — but there's no card on file yet."
topup_on_portal_to_continue: "Top up and manage billing on the portal to continue."
allowed_run_topup: "Remote Spending allowed. Run /topup to continue."
choice_resume_topup: "Resume {amount} top-up"
desc_resume_topup: "finish the held purchase"
resume_topup_title: "Resume your top-up"
resume_topup_detail: "{amount} is ready to finish — press Enter to resume."
resuming_topup: "Resuming your top-up — confirming settlement…"
# auto-reload (screen 4)
auto_reload: "Auto-reload"
auto_reload_hint: "Automatically add funds when your balance is low."
card_on_file: "Card on file: {card}"
no_saved_card_manage_portal: "No saved card — manage billing on the portal."
auto_reload_rule: "below {threshold} → reload to {reload_to}"
auto_reload_rule_cap: "Below {threshold} → reload to {reload_to}"
currently: "Currently: {rule}"
auto_reload_on_detail: "On — {rule}"
run_interactive_to_configure: "Run in the interactive CLI to configure auto-reload."
choice_edit_thresholds: "Edit thresholds"
desc_edit_thresholds: "change when / how much to reload"
choice_turn_off: "Turn off"
desc_turn_off: "disable auto-reload"
choice_agree_turn_on: "Agree and turn on"
desc_agree_turn_on: "enable auto-reload"
amount_prompt_labeled: "{label} (USD)"
threshold_label: "When balance falls below"
reload_to_label: "Reload balance to"
reload_must_exceed_threshold: "🔴 Reload-to amount must be greater than the threshold."
auto_reload_consent: "By confirming, you authorize Nous Research to charge {card} whenever your balance reaches {threshold}. Turn off any time here or on the portal."
turn_on_auto_reload_question: "Turn on auto-reload?"
auto_reload_enabled: "✅ Auto-reload on: below {threshold} → reload to {reload_to}."
auto_reload_disabled: "✅ Auto-reload turned off."
# monthly limit (screen 5)
monthly_spend_limit: "Monthly spend limit"
no_monthly_cap: "No monthly cap visible (managed on the portal)."
default_ceiling_suffix: " (default ceiling)"
used_this_month: "{spent} of {limit} used this month{ceiling}"
monthly_limit_read_only: "The monthly limit is set on the portal — the terminal shows it read-only."
subscription:
load_failed_label: "Could not load subscription"
team_header: "Team subscription"
a_team_org: "a team org"
team_connected: "This terminal is connected to {org}. Teams run on a shared balance · use /topup to add funds."
personal_note: "Personal subscriptions live on your personal account."
# overview status line
your_plan: "your plan"
cancels: "cancels"
end_of_billing_period: "the end of the billing period"
end_of_cycle: "the end of the cycle"
status_flip_cancels: " → cancels"
status_flip_to: " → {tier}"
status_left: " · ${amount} left"
status_view_only: " · view only"
status_renews: " · renews {date}"
status_line: "Plan: {plan}{flip}{left}{tail}"
status_free: "Plan: Free · free models only"
scheduled_change: "Scheduled change"
scheduled_change_line: "{from_plan} ──▶ {to_plan} {when}"
scheduled_when: "· {when}"
keep_until_then: "You keep {plan} (and its credits) until then."
paid_models_need_subscription: "> Paid models need a subscription. Start one to reach them."
low_balance: "! Low balance · {amount} left. Top up or upgrade before a mid-run cutoff."
plan_changes_need_admin: "Plan changes need an org admin/owner."
manage_url_line: "Manage your subscription: {url}"
open_then_rerun: "Open it in your browser, then re-run /subscription."
# free catalog
start_subscription: "Start a subscription"
choose_plan: "Choose a plan"
start_opens_portal: "Starting a subscription opens the portal to add your card."
desc_start_on_portal: "start {name} on the portal"
pick_plan_detail: "Pick a plan to open it on the portal."
cancelled_no_plan_started: "🟡 Cancelled. No plan started."
opening_portal_to_start: "Opening the portal to start {plan}…"
open_url_to_start: "Open this URL to start {plan}: {url}"
finish_in_browser: "Finish in your browser, then re-run /subscription."
# portal hand-off
manage_your_subscription: "Manage your subscription"
desc_open_subscription_page: "open the subscription page in your browser"
choice_copy_link: "Copy link"
desc_copy_link: "copy the manage-subscription URL to your clipboard"
copied: "📋 Copied: {url}"
manage_url: "Manage URL: {url}"
# change menu
choice_keep_plan: "Keep {plan} (undo the scheduled change)"
desc_keep_plan: "cancel the pending change"
choice_change_plan: "Change plan"
desc_change_plan: "upgrade or downgrade in the terminal"
choice_cancel_subscription: "Cancel subscription"
desc_cancel_subscription: "schedule cancellation at period end"
closed_no_change: "🟡 Closed. No plan change."
# tier picker
no_other_plans: "No other plans are available to switch to right now."
direction_upgrade: "upgrade"
direction_downgrade: "downgrade"
tier_row: "{row} · {direction}"
desc_switch_to: "switch to {name}"
change_plan_title: "Change plan"
current_pick_detail: "Current: {plan}. Pick a plan to preview the effect."
cancelled_no_plan_change: "🟡 Cancelled. No plan change."
# preview + confirm
checking_change: "Checking the change…"
the_selected_plan: "the selected plan"
already_on_nothing_to_change: "You are already on {plan} — nothing to change."
cannot_confirm_here: "This change cannot be confirmed here — manage it on the portal."
confirm_plan_change: "Confirm plan change"
tag_charged_now: "· charged now"
tag_scheduled_not_today: "· scheduled · not today"
charged_amount_now: "{amount} now (prorated)"
charged_prorated_now: "the prorated amount now"
upgrade_will_be_charged: "Upgrade to {plan}. You will be charged {charged}."
card_on_subscription_charged: "The card on your subscription will be charged."
named_card_on_subscription_charged: "{card} — the card on your subscription — will be charged."
choice_pay_and_upgrade: "Pay {amount} & upgrade now"
choice_upgrade_now_prorated: "Upgrade now (prorated charge)"
desc_charge_and_upgrade: "charge + upgrade now"
change_takes_effect: "Change to {plan} — takes effect {when}. No charge now; you keep your current plan until then."
choice_schedule_change: "Schedule change to {plan}"
desc_apply_change: "apply this change"
desc_do_not_change: "do not change"
monthly_credits_change: "Monthly credits change: {delta}."
# cancel
confirm_cancellation: "Confirm cancellation"
cancel_stays_active_until: "Cancel {plan} — it stays active until {end}, then won't renew."
keep_remaining_credits: "You keep your remaining credits for this period. You can resume before it ends."
cancel_subscription_question: "Cancel subscription?"
desc_keep_your_plan: "keep your plan"
cancelled_plan_unchanged: "🟡 Cancelled. Your plan is unchanged."
# apply outcomes
your_new_plan: "your new plan"
upgrade_already_on_tier: "You are already on {name}."
upgrade_upgraded: "Upgraded to {name}. Your new monthly credits land in a moment."
upgrade_requires_action: "🟡 This upgrade needs extra verification (3DS). Finish it on the portal."
upgrade_payment_failed: "🔴 Your card was declined. Update your payment method on the portal and try again."
pending_scheduled: "Scheduled — your plan doesn't change today. You keep it until the end of the billing period, then it switches."
pending_cancel_scheduled: "Scheduled — your plan stays active until the end of the billing period, then it cancels. Nothing changes today."
pending_resumed: "Undone — you stay on your current plan."
rerun_to_review: "Re-run /subscription anytime to review it."
stepup_explain: "To change your plan from the terminal, allow Remote Spending once. It opens your browser to authorize, then your change picks up right here."
stepup_noninteractive: "Run `hermes portal` and allow Remote Spending, then re-run /subscription."
stepup_declined: "No change made. Allow Remote Spending when you're ready."
stepup_not_granted: "Couldn't allow Remote Spending — an org admin or owner has to approve it for this org."
remote_spending_not_allowed_yet: "🟡 Remote Spending isn't allowed yet. Allow it, then retry."
upgrade_ambiguous: "🟡 Couldn't confirm the upgrade — your card may or may not have been charged."
rerun_to_check_plan: "Re-run /subscription to check your plan before trying again."
model:
# /model switch summary (labels shared with the gateway live under gateway.model.*)
switched: "✓ Model switched: {model}"
acting_model: "Acting model (billed for the run): {model}"
reasoning_effort: "Reasoning effort: {effort}"
reasoning_effort_saved: "Reasoning effort: {effort} (saved to config)"
saved: "Saved to config.yaml"
saved_global: "Saved to config.yaml (--global)"
next_turn_only: "(next turn only — restores after one response)"
session_only: "(session only — add --global to persist)"
# picker
effort_none: "none (disable reasoning)"
effort_keep_current: "Keep current effort"
no_authenticated_providers: "No authenticated providers found."
usage_switch: "switch model (this session)"
usage_switch_global: "switch model and persist as default"
usage_switch_once: "switch for the next turn only"
usage_switch_session: "switch for this session only"
usage_switch_provider_model: "switch provider + model"
usage_switch_reasoning: "switch and set reasoning effort"
usage_switch_provider: "switch provider"
usage_refresh: "re-fetch live model lists"
choice_switch_anyway: "Switch anyway"
desc_switch_anyway: "Use this model for the current Hermes session."
choice_cancel: "Cancel"
desc_keep_current_model: "Keep the current model."
switch_cancelled: "Model switch cancelled."
selection_failed: "✗ Model selection failed: {error}"
switch_failed: "⚠ Model switch to {model} failed ({error}); staying on {old_model}."
cleared_picker_cache: "Cleared model picker cache. Refreshing..."
moa_one_shot_queued: "MoA one-shot queued with preset {preset}; previous model will be restored after this turn."
# model-id normalization notices (yellow)
normalized_model: "Normalized model '{old}' to '{new}' for {provider}."
normalized_copilot_model: "Normalized Copilot model '{old}' to '{new}'."
stripped_provider_prefix: "Stripped provider prefix from '{old}'; using '{new}' for {provider}."
restored_from_session: "Model restored from session: {model}"
# openai_runtime sub-command
could_not_load_config: "❌ could not load config: {error}"
tip_reset_new_session: "Tip: `/reset` starts a new session immediately."
single_query:
query_label: "Query:"
kanban_run_reclaimed: "kanban: this worker's run was reclaimed before it started; exiting"
shared:
cancel: "Cancel"
back: "← Back"
unknown: "unknown"
unknown_error: "Unknown error"
noun_model: "model"
label_enabled: "enabled"
label_disabled: "disabled"
label_yes: "yes"
label_no: "no"
label_on: "on"
label_off: "off"
label_on_upper: "ON"
label_off_upper: "OFF"
label_value: "{label}: {value}"
just_now: "just now"
ago_seconds: "{seconds}s ago"
ago_minutes: "{minutes} min ago"
ago_hours: "{hours}h ago"
duration_s: "{seconds}s"
duration_m: "{minutes}m"
duration_ms: "{minutes}m {seconds}s"
duration_ms_compact: "{minutes}m{seconds}s"
duration_hm: "{hours}h {minutes}m"
duration_hms: "{hours}h {minutes}m {seconds}s"
duration_dhm: "{days}d {hours}h {minutes}m"
duration_m_rest: "{minutes}m {rest}"
duration_hm_rest: "{hours}h {minutes}m {rest}"
no_active_agent: "(._.) No active agent -- send a message first."
n_more: "{count} more"
more_not_shown: "more not shown"
yes_no_default_yes: "Y/n"
yes_no_default_no: "y/N"
yes_initial: "y"
tui:
approval_once: "Allow once"
approval_session: "Allow for this session"
approval_always: "Add to permanent allowlist"
approval_deny: "Deny"
approval_view: "Show full command"
approval_title: "⚠️ Dangerous Command"
approval_choose_full: "… (choose {label})"
approval_command_truncated: "… (command truncated — use /logs or /debug for full text)"
approval_description_truncated: "… (description truncated)"
confirm_title: "Confirm action"
confirm_footer: "Type 1/2/3 or use ↑/↓ then Enter. ESC/Ctrl+C cancels."
detail_truncated: "… (detail truncated)"
clarify_title: "{agent_name} needs your input"
clarify_question_count_one: "{count} question"
clarify_question_count_other: "{count} questions"
clarify_other_answer: "Other: {text}"
clarify_other_type_below: "Other (type below)"
clarify_other_type_answer: "Other (type your answer)"
clarify_guidance: "Type your answer in the prompt below, then press Enter."
picker_title_provider: "⚙ Model Picker — Select Provider"
picker_provider_row_one: "{name} ({count} model)"
picker_provider_row_other: "{name} ({count} models)"
picker_current_marker: " ← current"
picker_current_hint: "Current: {model} on {provider}"
picker_title_effort: "⚙ Model Picker — Reasoning effort for {model}"
picker_current_labeled: "{label} ← current"
picker_effort_hint: "Applies with the model switch (same scope) — Enter to choose"
picker_title_models: "⚙ Model Picker — {provider}"
picker_provider_fallback: "Provider"
picker_filter_hint: "Filter: {query}▏ ({shown}/{total} match — type to narrow, Backspace to clear)"
picker_select_hint: "Select a model ({count} available) — type to filter"
picker_empty: "No models listed for this provider. Use Back or Cancel."
palette_title: "⚙ Command Palette"
palette_filter_hint: "Filter: {query}▏ ({shown}/{total} match — Enter inserts, Esc cancels)"
palette_hint: "Type to filter {total} commands — ↑/↓ then Enter inserts, Esc cancels"
palette_empty: "(no matching commands)"
vault_code_title: "🔐 Verification code for {site}"
vault_code_body: "{site} is asking for a one-time code (text message, email or authenticator app).\nType the code and press Enter; {agent_name} enters it into the page for you.\nEnter on an empty line skips. The model never sees the code."
vault_save_title: "🔐 Save login for {site}"
vault_save_body: "The agent reached a sign-in page with no saved login for this site.\nType the email / username you sign in with (shown), then Enter.\nEnter on an empty line skips. Nothing here is shown to the model."
vault_save_pw_body: "Now the password (hidden). It is encrypted on this machine, bound to\n{origin}, and filled into the page without the model ever seeing it.\nEnter on an empty line skips."
vault_unlock_title: "🔐 Unlock {backend}"
vault_unlock_body: "The agent wants to sign into a site with a login saved in {backend}.\nType your master password (hidden) to unlock it for this session.\nEnter on an empty line keeps it locked. The model never sees the password."
sudo_title: "🔐 Sudo Password Required"
sudo_body: "Enter password below (hidden), or press Enter to skip"
secret_title: "🔑 Skill Setup Required"
secret_prompt: "Enter value for {var_name}"
secret_fallback_name: "secret"
secret_body: "Enter secret below (hidden), ESC or Ctrl+C to skip"
connection_connect: "Connect"
connection_continue: "Continue"
hint_sudo: "password hidden · Enter to skip"
hint_secret: "secret hidden · Enter to skip"
hint_approval: "↑/↓ to select, Enter to confirm"
hint_slash_confirm: "type 1/2/3, or ↑/↓ to select, Enter to confirm"
hint_connection_form: "type the value, Enter for next field · ↑/↓ move · ESC cancel"
hint_connection_url: "Enter to open in browser · ESC cancel"
hint_connection_authorized: "↑/↓ select, Enter to confirm"
hint_connection_waiting: "waiting for the backend… · Ctrl+C interrupt"
hint_connection_default: "connection setup"
hint_vault_username: "shown as you type · Enter to continue"
hint_clarify_freetext: "type your answer and press Enter"
hint_clarify_batch: "↑/↓ to select, Enter to lock, Tab next question"
hint_input_disabled: "input temporarily disabled"
hint_input_queues: "input stays active; Enter queues"
hint_command_running: "{frame} command in progress · {detail}"
placeholder_recording: "recording... {shortcut} to stop, Ctrl+C to cancel"
placeholder_transcribing: "transcribing..."
placeholder_vault_username: "type your email / username, Enter to continue · ESC to skip"
placeholder_vault_code: "type the code, Enter to submit · ESC to skip"
placeholder_password: "type password (hidden), Enter to submit · ESC to skip"
placeholder_secret: "type secret (hidden), Enter to submit · ESC to skip"
placeholder_choice: "type 1/2/3, or use ↑/↓ then Enter"
placeholder_answer: "type your answer here and press Enter"
placeholder_processing_command: "Processing command..."
placeholder_busy: "msg=interrupt · /queue · /bg · /steer · Ctrl+C cancel"
placeholder_voice_idle: "type or {shortcut} to record"
voice_recording_failed: "Voice recording failed: {error}"
recording_cancelled: "Recording cancelled."
force_exiting: "⚡ Force exiting..."
interrupting_again_hint: "⚡ Interrupting agent... (press Ctrl+C again to force exit)"
interrupting: "⚡ Interrupting agent..."
suspend_unsupported: "Suspend (Ctrl+Z) is not supported on Windows."
suspended: "{agent_name} has been suspended. Run `fg` to bring {agent_name} back."
steer_failed_queued: "Steer failed ({error}) — queued for next turn."
steered: "⏩ Steered: '{preview}'"
images_attached_preview_one: "[{count} image attached]"
images_attached_preview_other: "[{count} images attached]"
queued_next_turn: "Queued for the next turn: {preview}"
redirected_turn: "↪ Redirected current turn: '{preview}'"
model_selection_failed: "✗ Model selection failed: {error}"
auto_attached_image: "📎 Auto-attached image: {name}"
detected_file: "📄 Detected file: {name}"
images_attached_one: "📎 {count} image attached"
images_attached_other: "📎 {count} images attached"
command_interrupted: "Command interrupted."
welcome: "Welcome to Hermes Agent! Type your message or /help for commands."
activated_skills_label: "Activated skills:"
redaction_disabled_title: "⚠ Secret redaction is DISABLED"
redaction_disabled_body: "(HERMES_REDACT_SECRETS={value}). API keys and tokens may appear verbatim in chat output, session JSONs, and logs. Set {setting} in config.yaml to re-enable."
stdin_unavailable: "Error: stdin (fd 0) is not available.\nThis can happen with certain Python installations (e.g. uv-managed cPython on macOS).\nTry reinstalling Python via pyenv or Homebrew, then re-run: hermes setup"
shutting_down: "Shutting down… (finalizing session)"
session_deleted: "✓ Session {session_id} deleted"
session_delete_missing: "✗ Session {session_id} not found for deletion"
voice:
err_termux_app_missing: "Termux:API command package detected, but the Android app is missing.\nInstall/update the Termux:API Android app, then retry /voice on."
err_mic_access: "Voice mode needs microphone access.\nRun pkg install termux-api and install the Termux:API Android app, then retry /voice on."
err_audio_capture: "Voice mode requires audio capture."
err_stt_provider: "Voice mode requires an STT provider for transcription.\nRun hermes tools and configure Speech-to-Text, then restart Hermes."
silence_autostop: "Silence detected, auto-stopping..."
hint_continuous: "auto-stops on silence | {shortcut} to stop & exit continuous"
hint_termux: "Termux:API capture | {shortcut} to stop"
hint_stop: "{shortcut} to stop"
recording: "● Recording..."
autorestart_failed: "Voice auto-restart failed: {error}"
no_speech: "No speech detected."
preparing_stt: "Preparing local STT model '{model}' (first use may download it from Hugging Face)..."
transcribing: "Transcribing..."
stop_phrase: "Stop phrase detected — ending voice chat."
transcription_failed: "Transcription failed: {error}"
processing_error: "Voice processing error: {error}"
recording_preserved: "Recording preserved at: {path}"
no_speech_stop: "No speech detected 3 times, continuous mode stopped."
tts_failed: "TTS playback failed: {error}"
interjection: "🎤 Voice interjection — interrupting…"
echo_ignored: "Ignored likely TTS echo (not queued)."
already_enabled: "Voice mode is already enabled."
unavailable_header: "Voice mode unavailable in this environment:"
requirements_header: "Voice mode requirements not met:"
termux_install_cmd: "Run: pkg install termux-api"
termux_install_hint: "Then install/update the Termux:API Android app for microphone capture"
tts_enabled_suffix: " (TTS enabled)"
enabled: "Voice mode enabled{tts_status}"
record_key_hint: "{shortcut} to start/stop recording"
tts_toggle_hint: "/voice tts to toggle speech output"
disable_hint: "/voice off to disable voice mode"
stop_phrase_typed: "Stop phrase typed — ending voice chat."
disabled: "Voice mode disabled."
wake_unavailable: "Wake word unavailable: {error}"
wake_already: "Wake word is already listening."
wake_requirements_header: "Wake word requirements not met:"
wake_installing: "Installing wake word engine (first use — this may take a minute)..."
wake_start_failed: "Failed to start wake word: {error}"
wake_listening: "Wake word listening"
wake_listening_hint: "(say \"{phrase}\" — /wake off to stop)"
wake_stopped: "Wake word stopped."
wake_not_running: "Wake word is not running."
wake_profile_hint: "Wake phrase for profile '{profile}' — run: hermes -p {profile}"
wake_detected: "✦ Wake word detected — listening..."
wake_capture_failed: "Wake capture failed: {error}"
state_listening: "LISTENING"
state_paused: "PAUSED"
state_off: "OFF"
state_on: "ON"
state_yes: "YES"
wake_status_title: "Wake Word Status"
label_state: " State: {state}"
label_phrase: " Phrase: \"{phrase}\""
label_provider: " Provider: {provider}"
label_surface: " Surface: {surface}"
label_new_session: " New session: {value}"
mic_silent_warning: "⚠ Microphone delivers only silence — the listener can't hear anything."
mic_silent_macos_hint: "On macOS: System Settings > Privacy & Security > Microphone — allow your terminal/Hermes, then /wake off + /wake on."
wake_enable_hint: "Enable with /wake on"
enable_first_hint: "Enable voice mode first: /voice on"
tts_no_provider: "Warning: No TTS provider available. Install edge-tts or set API keys."
tts_status: "Voice TTS {status}."
status_title: "Voice Mode Status"
label_mode: " Mode: {value}"
label_tts: " TTS: {value}"
label_recording: " Recording: {value}"
label_record_key: " Record key: {shortcut}"
label_requirements: "Requirements:"
render:
response_transformed: "[Response transformed after streaming]"
banner_tagline: "{name} - AI Agent Framework"
banner_version: "Hermes Agent v{version} ({date})"
subagents:
cannot_steer_process: "Background processes cannot be steered; stop them or read the log."
count_live: "{count} live"
count_done: "{count} done"
count_running: "{count} running"
count_procs_one: "{count} proc"
count_procs_other: "{count} procs"
count_queued: "{count} queued"
goal_parked: "goal parked"
goal_paused: "goal paused"
goal_active: "goal active"
title_live_work: "Live work"
title_subagents: "Subagents"
title_processes: "Processes"
last_tool: "last: {tool}"
starting: "starting"
next_queued: "next: {text}"
heading_full: "{title} · {count} · Ctrl+T expand · Ctrl+R restore"
heading_medium: "{count} · Ctrl+T expand · Ctrl+R restore"
heading_compact: "{count} · Ctrl+T · Ctrl+R"
heading_session_restore: "{count} · Ctrl+R restore"
heading_session_compact: "{count} · Ctrl+R"
subagents_heading: "Subagents · {count} live · Ctrl+T expand · Ctrl+R collapse"
more_subagents: "+{count} more · Ctrl+T all subagents"
controls_expand_collapse: " · Ctrl+T expand · Ctrl+R collapse"
processes_heading: "Processes · {summary}{controls}"
more_processes: "+{count} more · Ctrl+T all processes"
controls_collapse: " · Ctrl+R collapse"
queue_heading: "Queue · {count} queued · /queue list{controls}"
more_queued: "+{count} more"
transcript_unavailable: "Live transcript not available yet."
steer_prompt: "Steer: "
roster_empty: "No live subagents or background processes. Results arrive in the conversation."
entry_gone: "This entry is no longer live."
noun_process: "process"
noun_subagent: "subagent"
footer_confirm_stop: "Stop selected {noun}? y confirm · Esc cancel"
footer_confirm_stop_compact: "Stop? y yes · Esc cancel"
footer_steering: "Enter queues guidance · Esc cancels (does not interrupt)"
footer_steering_compact: "Enter send · Esc cancel"
footer_steer_key: " s steer"
footer_steer_key_long: " · s steer"
footer_detail_compact: "PgUp/Dn ·{steer} x stop · Esc"
footer_roster_compact: "↑↓ · Enter tail · Ctrl+T close"
footer_detail: "Esc roster · PgUp/PgDn tail{steer} · x stop"
footer_roster: "↑/↓ select · Enter tail{steer} · x stop · q/Ctrl+T close"
dock:
process_goal_label: "background process"
last_output: "last: {detail}"
starting: "starting"
exit_code: "exit {code}"
verdict_killed: "killed"
verdict_lost: "lost"
verdict_ago: "{verdict} · {seconds}s ago"
process_gone: "This process is no longer tracked."
no_output: "No output yet."
approval:
bell_context: "approval"
label: "Approval"
timeout_denying: "⏱ Timeout — denying command"
timed_out_no_response: "timed out (no response)"
outcome_once: "allowed once"
outcome_session: "allowed for session"
outcome_always: "added to allowlist"
outcome_deny: "denied"
clarify:
bell_context: "clarify"
label: "Clarify"
timed_out_locked_answers: "(clarify timed out after {timeout}s — locked answers returned)"
connect:
bell_context: "connection setup"
cannot_restart: "This connection cannot be started again. Cancel and ask the agent again."
secret_set: "Set"
set_up: "Set up {name}"
button_connect: "Connect"
button_cancel: "Cancel"
button_continue: "Continue"
press_enter_open: "Press Enter to open in browser"
failed: "Connection failed"
authorized_no_tools: "Authorized. Tools unavailable."
connected: "Connected"
waiting: "Waiting…"
editor:
only_interactive: "External editor is only available inside the interactive CLI."
wait_for_command: "Wait for the current command to finish before opening the editor."
finish_active_prompt: "Finish the active prompt before opening the editor."
no_input_buffer: "No active input buffer is available for the external editor."
open_failed: "Failed to open external editor: {error}"
shell_command_failed: "Shell command failed: {error}"
command_failed: "Command failed: {error}"
queued_next_turn: "Queued for the next turn: {preview}"
modal:
cancelled_no_input: "🟡 /{command} cancelled (no input)."
unrecognized_choice: "🟡 Unrecognized choice '{raw}'. /{command} cancelled."
cancelled: "🟡 /{command} cancelled. {unchanged}"
re_enable_hint: " Re-enable via `approvals.{setting}: true` in config.yaml."
persist_opt_out_failed: "⚠️ Couldn't persist opt-out — {once_verb} once."
persist_failed_proceeding: "⚠️ Couldn't persist opt-out — proceeding once."
choice_prompt: "Choice [1/2/3]: "
destructive_title: "⚠️ /{command} — destroys conversation state"
approve_once: "Approve Once"
approve_once_detail: "proceed this time only"
always_approve: "Always Approve"
always_approve_detail: "proceed and silence this prompt permanently"
cancel: "Cancel"
cancel_detail: "keep current conversation"
conversation_unchanged: "Conversation unchanged."
destructive_always_msg: "🔒 Future /clear, /new, /reset, and /undo will run without confirmation."
bell_input_needed: "input needed"
bell_turn_complete: "turn complete"
palette:
category_skill: "Skill"
secret:
bell_sudo: "sudo password"
sudo_timeout: "⏱ Timeout — continuing without sudo"
sudo_received: "✓ Password received (cached for session)"
sudo_skipped: "⏭ Skipped"
bell_unlock: "unlock {name}"
stays_locked: "⏭ {name} stays locked"
unlocking_for_session: "✓ Unlocking {name} for this session"
bell_save_login: "save login for {site}"
login_not_saved: "⏭ Not saving a login for {site}"
login_saved: "✓ Login for {site} saved to your vault"
bell_verification_code: "verification code for {site}"
code_not_entered: "⏭ No code entered for {site}"
code_entered: "✓ Code entered into {site}"
display:
ui_redrawn: "✓ UI redrawn"
status_bar_visible: "Status bar visible"
status_bar_hidden: "Status bar hidden"
exit:
unknown_argument: "✗ Unknown argument: {argument}. Use /exit --delete to also remove session history."
goal:
barrier_lifted: "▶ Goal barrier lifted — resuming."
paused_interrupted: "⏸ Goal paused — turn was interrupted. Use /goal resume to continue, or /goal clear to stop."
loop:
wakeup_firing: "↻ /loop wakeup #{tick} firing…"
paused_interrupted: "⏸ Loop paused — wakeup turn was interrupted. Use /loop resume to continue, or /loop stop to end it."
remaining: "↻ Loop: {label}."
plugins:
none_installed: "No user plugins installed."
install_hint: "Install one: hermes plugins install owner/repo"
drop_dir_hint: "Or drop a plugin directory into {path}"
bundled_available: "({count} bundled plugins available — see: hermes plugins list)"
user_plugins_header: "User plugins ({count}):"
count_noun_tools: "tools"
count_noun_hooks: "hooks"
count_noun_commands: "commands"
bundled_more: "(+{count} bundled — see: hermes plugins list)"
enable_disable_hint: "Enable/disable: hermes plugins enable/disable <name>"
system_error: "Plugin system error: {error}"
queue:
usage: "Usage: /queue <prompt> | /queue list | /queue edit N <prompt> | /queue rm N | /queue move FROM TO | /queue clear"
queued: "Queued: {preview}"
queued_next_turn: "Queued for the next turn: {preview}"
empty: "Queue is empty."
pending_header: "Queue ({count} pending):"
cleared_one: "Cleared {count} queued prompt."
cleared_other: "Cleared {count} queued prompts."
removed: "Removed queue item {index}: {preview}"
item_not_found: "Queue item {index} not found. Current size: {size}"
usage_edit: "Usage: /queue edit <number> <new prompt>"
updated: "Updated queue item {index}: {preview}"
usage_move: "Usage: /queue move <from> <to>"
move_out_of_range: "Queue move out of range. Current size: {size}"
moved: "Moved queue item {source} to {destination}."
usage_add: "Usage: /queue add <prompt>"
reload:
env_reloaded: "Reloaded .env ({count} var(s) updated)"
steer:
usage: "Usage: /steer <prompt>"
failed: "Steer failed: {error}"
queued: "⏩ Steer queued — arrives after the next tool call: {preview}"
rejected_empty: "Steer rejected (empty payload)."
no_agent_queued: "No agent running; queued as next turn: {preview}"
title:
session_id: "Session ID: {session_id}"
current: "Title: {title}"
pending: "Title (pending): {title}"
none_set_usage: "No title set. Usage: /title <your session title>"
usage: "Usage: /title <your session title>"
empty_after_cleanup: "Title is empty after cleanup. Please use printable characters."
set: "Session title set: {title}"
already_in_use: "Title '{title}' is already in use by session {session_id}"
queued: "Session title queued: {title} (will be saved on first message)"
undo:
invalid_count: "(._.) Invalid count {value} — use /undo or /undo N."
removes_last_exchange: "This removes the last user/assistant exchange from history."
removes_last_turns: "This removes the last {count} user turns from history."
resume:
tool_calls_one: "[{count} tool call: {names}]"
tool_calls_other: "[{count} tool calls: {names}]"
event_model_changed: "model changed"
event_delegation_completed: "background delegation completed"
event_process_finished: "background process finished"
event_resumed_interrupted: "resumed interrupted turn"
session_id_hint: "Use a session ID from a previous CLI run (hermes sessions list)."
session_not_found: "Session not found: {session_id}"
compressed_into: "Session {session_id} was compressed into {descendant}; resuming the descendant with your transcript."
cannot_resume: "Cannot resume session: {error}"
cannot_resume_prefix: "Cannot resume session:"
resumed_session: "↻ Resumed session"
counts_one: "({count} user message, {total} total messages)"
counts_other: "({count} user messages, {total} total messages)"
session_no_messages: "Session {session_id} found but has no messages. Starting fresh."
title_applied: "Session title applied: {title}"
title_apply_failed: "Could not apply pending title: {error}"
label_you: "● You:"
label_assistant: "◆ {agent_name}:"
earlier_messages: "... {count} earlier messages ..."
panel_title: "Previous Conversation"
startup:
cooldown_cause_rate_limit: "after a rate-limit or quota response"
cooldown_cause_token_refresh: "after a failed token refresh"
cooldown_cause_http: "after an HTTP {code} response"
cooldown_cause_failed_request: "after a failed request"
credential_cooling_down: "The {provider} credential is cooling down {cause}; it re-enters rotation in about {minutes}m."
sign_in_lost: "sign-in lost"
sign_in_was_lost: "The {provider} sign-in was lost ({reason}); run `hermes auth add {provider}` to sign in again."
provider_resolution_failed: "Provider resolution failed."
no_api_key_for_provider: "⚠️ No API key found for provider '{provider}'."
no_provider_configured: "⚠️ No inference provider is configured."
run_model_or_setup: "Run 'hermes model' to choose a provider, or 'hermes setup' for first-time setup."
empty_base_url: "⚠️ Provider resolver returned an empty base URL. Check your provider config or run: hermes setup"
switching_to_fallback: "⚠️ {reason} — switching to fallback: {provider} / {model}"
first_run_no_provider: "☤ No inference provider is configured yet — let's fix that."
first_run_pick_provider: "You'll pick a provider (Nous Portal OAuth is the fastest; no API key needed) and a model."
first_run_prompt: "Set up a provider now? [Y/n]:"
first_run_skipped: "Skipped. Run 'hermes model' or 'hermes setup' any time."
first_run_cancelled: "Setup cancelled. Run 'hermes model' any time."
provider_setup_failed: "⚠️ Provider setup failed: {error}"
run_model_to_retry: "Run 'hermes model' to try again."
provider_configured: "✓ Provider configured — you're ready to chat."
provider_setup_incomplete: "Provider setup didn't complete. Run 'hermes model' to retry."
tirith_unavailable: "⚠ tirith security scanner enabled but not available — command scanning will use pattern matching only"
unknown_toolsets: "Warning: Unknown toolsets: {names}"
stdin_unusable: "\nError: stdin is not usable ({error}).\nThis can happen with certain Python installations (e.g. uv-managed cPython on macOS)\nwhere kqueue cannot register fd 0.\nTry reinstalling Python via pyenv or Homebrew, then re-run: hermes setup"
warning: "Warning: {warning}"
chat:
initializing_agent: "Initializing agent..."
context_refs: "[@ context: {count} ref(s), {tokens} tokens]"
context_injection_refused: "Context injection refused."
skipped_unreadable_images: "⚠ skipped {count} unreadable image path(s)"
attaching_images_natively: "📎 attaching {count} image(s) natively (model supports vision): {names}"
new_message_interrupting: "⚡ New message detected, interrupting..."
unknown_error: "Unknown error"
continuous_voice_stopped: "Continuous voice mode stopped due to error."
interrupted_marker: "── [Interrupted — processing new message] ──"
iteration_budget_reached: "⚠ Iteration budget reached ({used}/{max}) — response may be incomplete"
sending_after_interrupt_multi: "⚡ Sending {count} messages after interrupt: '{preview}'"
sending_after_interrupt: "⚡ Sending after interrupt: '{preview}'"
delivering_leftover_steer: "⏩ Delivering leftover /steer as next turn: '{preview}'"
reasoning_label: "Reasoning"
reasoning_more_lines: "... ({count} more lines — /reasoning full to show)"
billing_cta_nous: "Run [bold]/topup[/] to add credits, or [bold]/subscription[/] to change plan."
billing_cta_provider: "Add credits with {provider}"
your_provider: "your provider"
billing_cta_switch: "Or switch providers with [bold]/model <model> --provider <provider>[/]."
out_of_credits_title: "⚡ Out of credits"
session:
fresh_start: "✨ (◕‿◕)✨ Fresh start! Screen cleared and conversation reset."
model_reset_default: "(model reset to config default: {model})"
title_rejected: "Title rejected: {error}"
title_empty_untitled: "Title is empty after cleanup — session started untitled."
title_error_untitled: "{error} — session started untitled."
cwd_gone: "⚠ Session's working directory is gone: {recorded} — staying in {current}"
cwd_restored: "↻ Working directory: {path}"
cwd_enter_failed: "⚠ Could not enter session's working directory {path}: {error}"
yolo_restored: "⚡ YOLO mode restored from session — all commands auto-approved. /yolo to turn off."
invalid_checkpoint: "Invalid checkpoint number. Use 1-{count}."
status_tools_deferred: "tools deferred"
status_tools_count: "{count} tools"
status_toolsets: "toolsets: {toolsets}"
status_provider: "provider: {provider}"
status_auth: "auth: {source}"
reasoning_display_suffix: " (display: {state})"
yolo_bypass_suffix: " (YOLO bypass active)"
ctx_left_prefix: "{percent}% left · "
ctx_tokens_used: "{left}{used} / {max} tokens used"
status_title: "Hermes CLI Status"
label_reasoning: "Reasoning"
label_approvals: "Approvals"
label_context: "Context"
recent_intro: "(._.) No messages in the current chat yet — here are recent sessions you can resume:"
recent_header: "Recent sessions:"
column_title: "Title"
column_preview: "Preview"
column_last_active: "Last Active"
column_id: "ID"
resume_usage: "Use /resume <number>, /resume <session id>, or /resume <session title> to continue."
resume_example: "Example: /resume 2"
history_empty: "(._.) No conversation history yet."
history_tools_header: "[Tools]"
history_tools_hidden_one: "({count} tool message hidden)"
history_tools_hidden_other: "({count} tool messages hidden)"
history_title: "(^_^) Conversation History"
history_you_header: "[You #{index}]"
history_assistant_header: "[{agent_name} #{index}]"
history_no_text: "(no text response)"
history_tool_calls_one: "(requested {count} tool call)"
history_tool_calls_other: "(requested {count} tool calls)"
new_session_titled: "(^_^)v New session started: {title}"
new_session: "(^_^)v New session started!"
resume_index_out_of_range: "Resume index {index} is out of range."
resume_no_args_hint: "Use /resume with no arguments to see available sessions."
save_nothing: "(;_;) No conversation to save."
save_dir_failed: "(x_x) Failed to create save directory {path}: {error}"
format_markdown: "Markdown"
save_ok: "(^_^)v Conversation saved to: {path} ({label})"
save_resume_hint: "Resume the live session with: hermes --resume {session_id}"
save_failed: "(x_x) Failed to save: {error}"
retry_no_messages: "(._.) No messages to retry."
retry_no_user_message: "(._.) No user message found to retry."
retry_unsafe: "(._.) Cannot retry that message safely: {error}"
retry_rewind_failed: "(x_x) Retry rewind failed; history was not changed: {error}"
retrying: "(^_^)b Retrying: \"{preview}\""
undo_no_messages: "(._.) No messages to undo."
undo_no_user_message: "(._.) No user message found to undo."
undo_failed: "(x_x) Undo failed; history was not changed: {error}"
undo_ok_one: "(^_^)b Undid {count} turn ({messages} message(s)). Backed up to: \"{backup}\""
undo_ok_other: "(^_^)b Undid {count} turns ({messages} message(s)). Backed up to: \"{backup}\""
undo_remaining: "{count} message(s) remaining in history."
yolo_state_locked_on: "locked ON"
yolo_locked: "⚡ YOLO is {state} for this process (started with --yolo / HERMES_YOLO_MODE). /yolo cannot disable it — restart without the flag to re-enable approvals."
yolo_off: "⚠ YOLO mode {state} — dangerous commands will require approval."
yolo_on: "⚡ YOLO mode {state} — all commands auto-approved. Use with caution."
compress_too_short: "(._.) Not enough conversation to compress (need at least {min} messages)."
compress_no_agent: "(._.) No active agent -- send a message first."
compress_busy: "Compressing context..."
compress_summarizing: "🗜️ Summarizing up to here: {count} messages, keeping last {keep_last} exchange(s) verbatim..."
compress_focus: "🗜️ Compressing {count} messages, focus: \"{focus}\"..."
compress_plain: "🗜️ Compressing {count} messages..."
compress_failed: "❌ Compression failed: {error}"
persist_prompt_summary: "{icon} {label}: {detail} → {outcome}"
goodbye: "Goodbye! ☤"
exit_resume_hint: "Resume this session with:"
exit_label_session: "Session: {session_id}"
exit_label_title: "Title: {title}"
exit_label_duration: "Duration: {duration}"
exit_label_messages: "Messages: {count} ({user} user, {tool_calls} tool calls)"
status_bar:
vim_insert: "INSERT"
vim_replace: "REPLACE"
vim_normal: "NORMAL"
battery_none: "no battery detected on this machine"
battery_none_nothing_shown: "no battery detected, so nothing will show here"
battery_currently: " — currently {reading}"
battery_state: "Battery indicator {state}{detail}"
battery_usage: "Usage: /battery [on|off|status]"
battery_on: "Battery indicator on{detail}"
battery_off: "Battery indicator off"
latency_label: "{value}s"
velocity_label: "{value} t/s"
voice_rec: "● REC"
voice_rec_stop_hint: "● REC {shortcut} to stop"
voice_stt: "◉ STT"
voice_transcribing: "◉ Transcribing..."
voice_tts_on: " | TTS on"
voice_continuous: " | Continuous"
voice_mode_hint: "🎤 Voice mode{tts}{cont} — {shortcut} to record"
goal_turns: "⊙ goal {used}/{max}"
goal: "⊙ goal"
ctx_pinned: " pinned"
ctx_unknown: "ctx --"
yolo_badge: "⚠ YOLO"
stash_header_one: "📌 Stash ({count} item)"
stash_header_other: "📌 Stash ({count} items)"
stash_footer: "↑↓ Enter=restore D=delete Esc"
stash_footer_compact: "↑↓ ⏎ D Esc"
stream:
busy_skills_search: "Searching skills..."
busy_skills_browse: "Loading skills..."
busy_skills_inspect: "Inspecting skill..."
busy_skills_install: "Installing skill..."
busy_skills: "Processing skills command..."
busy_browser: "Configuring browser..."
busy_reload_mcp: "Reloading MCP servers..."
busy_reload_skills: "Reloading skills..."
processing_command: "Processing command..."
thinking_label: "[thinking]"
thinking_more_lines: "... ({count} more lines)"
preview_more_lines_one: "... (+{count} more line)"
preview_more_lines_other: "... (+{count} more lines)"
vision_analyzing: "👁️ analyzing {name} ({size_kb}KB)..."
vision_analyzed: "✓ image analyzed"
vision_failed: "⚠ vision analysis failed — path included for retry"
vision_error: "⚠ vision analysis error — path included for retry"
tool_preparing: "{emoji} preparing {what}…"
reference_noun: "reference"
reference_header_indexed: "Reference {index}/{count} — {label}"
reference_header: "Reference — {label}"
aggregating_detail: "◆ aggregating ({detail})"
aggregating: "◆ aggregating"
background_running_one: "↩ Background task running — I'll resume when it finishes. Keep chatting."
background_running_other: "↩ Background {count} tasks running — I'll resume when they finish. Keep chatting."
terminal:
recovered_mouse_modes: "Recovered terminal input modes after leaked mouse reports. If this repeats, run /new or restart this tab."
recovered_cooked_mode: "Recovered terminal from cooked-mode drift (input should respond normally again)."
image_badge_count: "[📎 {count} images attached]"
image_badge_numbered: "[📎 Image #{index}]"
commands:
shared:
more_suffix: " (+{count} more)"
set_saved: "✓ {label} set to '{value}' (saved to config)"
set_session_only: "✓ {label} set to '{value}' (session only)"
scope_saved: "(saved to config)"
scope_save_failed: "(session only; config save failed)"
scope_session: "(this session — use --global to persist)"
unknown_argument: "(._.) Unknown argument: {arg}"
usage: "Usage: {usage}"
needs_active_session: "{command} needs an active chat session to run."
unavailable_no_session: "{label} unavailable (no active session)."
state_on: "ON"
state_off: "OFF"
error: "Error: {error}"
agent_busy: "Agent is busy. Wait for the current turn to finish, then retry {command}."
no_active_agent_session: "No active agent session."
could_not_generate_diff: "Could not generate diff"
busy:
short_queue: "queues for next turn"
short_steer: "steers into current run (after next tool call)"
short_interrupt: "redirects current run immediately"
long_queue: "Enter will queue follow-up input while Hermes is busy."
long_steer: "Enter will steer your message into the current run (after the next tool call)."
long_interrupt: "Enter will redirect the current run while Hermes is busy; /stop still cancels it."
usage: "Usage: /busy [queue|steer|interrupt|status]"
status: "Busy input mode: {mode}"
enter_while_busy: "Enter while busy: {behavior}"
label: "Busy input mode"
reasoning:
headline_on: "ON"
headline_off: "OFF"
headline_full: "FULL"
headline_clamped: "CLAMPED to 10 lines"
note_show: "Model thinking will be shown during and after each response."
note_full: "The post-response recap box will print complete thinking."
display_on: "on ✓"
display_off: "off"
full_state_full: "full"
full_state_clamped: "clamped to 10 lines"
effort_line: "Reasoning effort: {level}"
display_line: "Reasoning display: {display} ({full})"
usage: "Usage: /reasoning <none|minimal|low|medium|high|xhigh|max|ultra|show|hide|full|clamp> [--global]"
display_saved: "✓ Reasoning display: {headline} (saved)"
display_off_note: "Note: reasoning display is OFF — run /reasoning show to see it."
valid_levels: "Valid levels: none, minimal, low, medium, high, xhigh, max, ultra"
valid_display: "Display: show, hide"
valid_scope: "Scope: session-scoped by default, --global to persist"
effort_set: "✓ Reasoning effort set to '{effort}' {scope}"
indicator:
usage: "Usage: /indicator [{styles}]"
status: "Busy-indicator style: {style}"
unknown_style: "(._.) Unknown indicator style: {arg}"
label: "Busy-indicator style"
note: "The TUI picks up the new style on its next render."
fast:
not_supported: "(._.) /fast is only available for models that support fast mode (OpenAI Priority Processing or Anthropic Fast Mode)."
feature_generic: "Fast mode"
feature_anthropic: "Anthropic Fast Mode"
feature_openai: "Priority Processing"
feature_ultrafast: "OpenAI Ultrafast"
ultrafast_not_supported: "(._.) Ultrafast is only available for GPT-6 Astra (current model: {model})."
usage: "Usage: /fast [normal|fast|auto|cold|ultrafast|status] [--global]"
status: "{feature}: {status}"
set_to: "✓ {feature} set to {value} {scope}"
hatch:
progress_compose: "┊ composing spritesheet…"
progress_save: "┊ saving…"
progress_drawing: "┊ drawing {state}…"
describe_prompt: "(o_o) Describe your pet: "
usage: "(o_o) Usage: /hatch <description> (e.g. /hatch a tiny cyber fox)"
default_name: "Pet"
designing: "(o_o) Designing '{concept}'… (a minute of image-model calls)"
base_failed: "(x_x) Couldn't generate a base look: {error}"
no_base_draft: "(x_x) No base draft came back — try again."
failed: "(x_x) Hatch failed: {error}"
hatched: "(^_^)b {name} hatched and adopted — it'll pop in shortly!"
pet:
is_out: "(^_^)b {name} is out — it'll pop in shortly."
put_away_named: "(-_-)zzZ {name} put away."
put_away: "(-_-)zzZ Pet put away."
usage_scale: "(o_o) Usage: /pet scale <factor> (e.g. /pet scale 0.5)"
scale_set: "(^_^) Pet scale → {scale}."
fetching: "(o_o) Fetching '{name}' from petdex…"
adopt_failed: "(x_x) Couldn't adopt '{name}': {error}"
error: "(x_x) {error}"
diff:
label_working: "Unstaged"
label_staged: "Staged"
label_all: "All (vs HEAD)"
untracked: "Untracked:"
and_more: "... and {count} more"
more_lines: "... ({count} more lines — {hint})"
stat_hint: "run /diff --stat for a summary"
stat_hint_session: "run /diff session --stat for a summary"
not_enabled: "Checkpoints are not enabled, so there's no session baseline.\nEnable with: hermes --checkpoints\nOr in config.yaml: checkpoints: { enabled: true }\n(Plain /diff still works — it uses git directly.)"
no_session_changes: "No changes — Hermes hasn't edited any files here yet."
rollback:
not_enabled: "Checkpoints are not enabled.\nEnable with: hermes --checkpoints\nOr in config.yaml: checkpoints: { enabled: true }"
none_here_showing_all: "No checkpoints for {cwd} — showing all directories."
all_directories: "all directories"
usage_diff: "Usage: /rollback diff <N>"
no_changes_since: "No changes since this checkpoint."
more_lines: "... ({count} more lines, showing first 80)"
restored_all: "✅ Restored to checkpoint {hash}: {reason}"
restored_file: "✅ Restored {file} from checkpoint {hash}: {reason}"
snapshot_saved: "A pre-rollback snapshot was saved automatically."
chat_turn_undone: "Chat turn undone to match restored file state."
snapshot:
unknown_subcommand: "Unknown subcommand: {subcommand}"
usage: "Usage: /snapshot [list|create [label]|restore <id>|prune [N]]"
none_yet: "No state snapshots yet."
create_hint: "Create one: /snapshot create [label]"
list_header: "State snapshots ({dir}/state-snapshots/):"
col_num: "#"
col_id: "ID"
col_files: "Files"
col_size: "Size"
col_label: "Label"
created: "Snapshot created: {snapshot_id}"
nothing_to_snapshot: "No state files found to snapshot."
usage_restore: "Usage: /snapshot restore <snapshot-id>"
most_recent: "Most recent: {snapshot_id}"
invalid_number: "Invalid snapshot number. Use 1-{max}."
restored: "Restored state from: {snapshot_id}"
restart_recommended: "Restart recommended for gateway/dashboard processes to pick up state.db changes."
not_found: "Snapshot not found: {snapshot_id}"
usage_prune: "Usage: /snapshot prune [keep-count]"
pruned: "Pruned {deleted} old snapshot(s) (keeping {keep})."
export:
usage: "Usage: /export [profile] [-o output.tar.gz]"
exported: "✓ Exported '{name}' to {path}"
share_hint: "Share it: the other user runs /import or `hermes profile import <archive>`."
import:
usage: "Usage: /import <archive.tar.gz> [--name <name>]"
imported: "✓ Imported profile '{name}' at {path}"
wrapper_created: "Wrapper created: {path}"
use_it: "Use it: hermes -p {name}"
stop:
none_running: "No running background processes."
stopping: "Stopping {count} background process(es)..."
stopped: "✅ Stopped {count} process(es)."
interrupted_delegations: "✅ Interrupted {count} background delegation(s)."
agents:
running_processes: "Running processes: {count}"
recently_finished: "Recently finished: {count}"
background_delegations: "Background delegations: {count} running"
no_progress: " · no progress {seconds}s — interrupting"
quiet: " · quiet {seconds}s"
in_tool: "in {tool}"
between_turns: "between turns"
child_line: "└ child {index}: {api_calls} api calls · {doing}"
last_activity: " · last activity {seconds}s ago"
agent_running: "Agent: running"
agent_idle: "Agent: idle"
journey:
failed: "/journey failed: {error}"
paste:
no_image: "(._.) No image found in clipboard"
attached: "📎 Image #{index} attached from clipboard"
extract_failed: "(>_<) Clipboard has an image but extraction failed"
copy:
nothing_yet: "Nothing to copy yet."
usage: "Usage: /copy [number]"
invalid_number: "Invalid response number. Use 1-{max}."
nothing_in_responses: "Nothing to copy in assistant responses yet."
nothing_in_response: "Nothing to copy in that assistant response."
copied_osc52: "Copied assistant response #{index} via OSC 52 (terminal support required)"
copied: "Copied assistant response #{index} to clipboard"
failed: "Clipboard copy failed: {error}"
image:
usage: "Usage: /image <path> e.g. /image {example}"
not_found: "(>_<) File not found: {path}"
unsupported: "(._.) Not a supported image file: {name}"
attached: "📎 Attached image: {name}"
now_type_prompt: "Now type your prompt (or use --image in single-query mode): {text}"
tools:
usage: "(._.) Usage: /tools {subcommand} <name> [name ...]"
example_toolset: "Built-in toolset: /tools {subcommand} web"
example_mcp: "MCP tool: /tools {subcommand} github:create_issue"
disabling: "Disabling {names}..."
enabling: "Enabling {names}..."
session_reset: "Session reset. New tool configuration is active."
profile:
profile: "Profile: {profile}"
home: "Home: {home}"
handoff:
usage: "Usage: /handoff <platform>"
usage_detail_1: "Hands the current session off to that platform's home channel."
usage_detail_2: "The CLI session ends here; resume it later with /resume."
already_in_flight: "Session is already in flight for handoff. Wait for it to settle, then retry."
queued: "Queued handoff of '{title}' → {platform} (home: {home})."
waiting: "Waiting for the gateway to pick it up..."
config_load_failed: "Could not load gateway config: {error}"
unknown_platform: "Unknown platform '{platform}'."
platform_not_enabled: "Platform '{platform}' is not configured/enabled in the gateway."
no_home_channel: "No home channel configured for {platform}."
set_home_hint: "Set one with /sethome on the destination chat first."
session_row_failed: "Could not ensure session row in state.db: {error}"
picked_up: "Gateway picked it up; transferring..."
complete: "↻ Handoff complete. The session is now active on {platform}."
resume_hint: "Resume it on this CLI later with: /resume {title}"
unknown_error: "unknown error"
failed: "Handoff failed: {error}"
session_intact_retry: "Your CLI session is intact. Try /handoff again, or /resume on the platform manually."
still_transferring: "Still transferring (the agent is replaying your session on the destination)..."
taking_long: "The gateway is taking unusually long to finish the transfer."
check_platform: "Check {platform} — the session may still arrive there."
no_longer_waiting: "This CLI is no longer waiting. Avoid continuing this session here;"
retry_when_settled: "if nothing arrives, retry /handoff once the state settles."
timed_out: "Timed out waiting for the gateway. Is `hermes gateway` running?"
session_intact: "Your CLI session is intact."
resume:
usage: "Usage: /resume <number|session_id_or_title>"
tip_find_sessions: "Tip: Use /history or `hermes sessions list` to find sessions."
already_on: "Already on that session."
resumed_one: "↻ Resumed session {session_id}{title} ({count} user message, {total} total)"
resumed_other: "↻ Resumed session {session_id}{title} ({count} user messages, {total} total)"
resumed_empty: "↻ Resumed session {session_id}{title} — no messages, starting fresh."
not_found: "Session not found: {target}"
not_found_hint: "Use /sessions or `hermes sessions list` to see available sessions."
compressed_redirect: "Session {session_id} was compressed into {resolved_id}; resuming the descendant with your transcript."
sessions:
none_yet: "(._.) No previous sessions yet."
branch:
branched_one: "⑂ Branched session \"{title}\" ({count} user message)"
branched_other: "⑂ Branched session \"{title}\" ({count} user messages)"
original_session: "Original session: {session_id}"
branch_session: "Branch session: {session_id}"
worktree:
active: "Active worktree: {path}"
branch: "Branch: {branch}"
none_active: "No active worktree for this session."
hint_new: "/worktree new [name] — create one and move this session into it"
hint_prune: "/worktree prune — reclaim stale trees and merged branches"
not_in_repo_note: "(not inside a git repository)"
unknown_subcommand: "Unknown /worktree subcommand: {subcommand}"
usage: "Usage: /worktree [new [name] | list]"
new_requires_repo: "❌ /worktree new requires being inside a git repository."
not_in_repo: "Not inside a git repository."
actions_planned: "{count} action(s) planned."
actions_done: "{count} action(s) done."
nothing_to_reclaim: "Nothing to reclaim — remaining trees/branches carry real work."
preserved: "Preserved {count} tree(s) with real work:"
list_failed: "Could not list worktrees."
enter_failed: "⚠ Created worktree but could not enter it: {error}"
ready: "✅ Worktree ready: {path}"
tools_in_worktree: "Terminal and file tools now operate in the worktree."
personality:
title: "(^o^)/ Personalities"
none_option: "(no personality overlay)"
usage: "Usage: /personality <name> (* = active)"
unknown: "(._.) Unknown personality: {name}"
available: "Available: none, {available}"
scope_saved: "(saved to config)"
scope_session: "(session only)"
cleared: "{face} Personality cleared {scope}"
cleared_note: "No personality overlay — using base agent behavior."
set_to: "{face} Personality set to '{name}' {scope}"
cron:
repeat_must_be_integer: "(._.) --repeat must be an integer"
unknown_command: "(._.) Unknown cron command: {subcommand}"
available: "Available: list, add, edit, pause, resume, run, remove"
overview_title: "(^_^) Scheduled Tasks"
overview_commands: "Commands:"
overview_usage: "/cron list\n/cron add \"every 2h\" \"Check server status\" [--skill blogwatcher]\n/cron edit <job_id> --schedule \"every 4h\" --prompt \"New task\"\n/cron edit <job_id> --skill blogwatcher --skill maps\n/cron edit <job_id> --remove-skill blogwatcher\n/cron edit <job_id> --clear-skills\n/cron pause <job_id>\n/cron resume <job_id>\n/cron run <job_id>\n/cron remove <job_id>"
current_jobs: "Current Jobs:"
skills: "Skills: {skills}"
skills_none: "Skills: none"
next_short: "Next"
no_jobs_hint: "No scheduled jobs. Use '/cron add' to create one."
no_jobs: "(._.) No scheduled jobs."
list_header: "Scheduled Jobs:"
id: "ID: {job_id}"
name: "Name: {name}"
state: "State: {state}"
schedule: "Schedule: {schedule}"
schedule_repeat: "Schedule: {schedule} ({repeat})"
next_run: "Next run: {value}"
next_run_na: "Next run: N/A"
prompt: "Prompt: {prompt}"
status_delivery_failed: "delivery_failed: {error}"
status_error: "error: {error}"
last_run: "Last run: {when} ({status})"
usage_add: "(._.) Usage: /cron add <schedule> <prompt>"
need_prompt_or_skill: "(._.) Please provide a prompt or at least one skill"
create_failed: "(x_x) Failed to create job: {error}"
created: "(^_^)b Created job: {job_id}"
usage_edit: "(._.) Usage: /cron edit <job_id> [--schedule ...] [--prompt ...] [--skill ...]"
job_not_found: "(._.) Job not found: {job_id}"
update_failed: "(x_x) Failed to update job: {error}"
updated: "(^_^)b Updated job: {job_id}"
usage_action: "(._.) Usage: /cron {subcommand} <job_id>"
pause_failed: "(x_x) Failed to pause job: {error}"
resume_failed: "(x_x) Failed to resume job: {error}"
run_failed: "(x_x) Failed to run job: {error}"
remove_failed: "(x_x) Failed to remove job: {error}"
removed: "(^_^)b Removed job: {name} ({job_id})"
did_not_run: "(x_x) Did not run job: {name} ({job_id})\n {reason}"
paused: "(^_^)b Paused job: {name} ({job_id})"
resumed: "(^_^)b Resumed job: {name} ({job_id})"
triggered: "(^_^)b Triggered job: {name} ({job_id})"
suggestions:
failed: "Suggestions command failed: {error}"
blueprint:
failed: "Cron blueprint command failed: {error}"
curator:
failed: "(._.) curator: {error}"
kanban:
failed: "(._.) kanban error: {error}"
memory:
unknown_subcommand: "Unknown /memory subcommand. Use: pending, approve <id>, reject <id>, approval <on|off>."
learn:
from_description: "⚡ Learning a skill from what you described..."
from_conversation: "⚡ Learning a skill from this conversation..."
plan:
planning: "📋 Planning: {task}"
from_conversation: "📋 Planning from this conversation's context..."
init:
updating: "⚡ Updating AGENTS.md from a project scan..."
generating: "⚡ Generating AGENTS.md from a project scan..."
background:
usage: "Usage: /bg <prompt>\nExample: /bg Summarize the top HN stories today\n(For a side question about this conversation, use /btw <question>.)\nThe task runs in a separate session and results display here when done."
no_credentials: "(>_<) Cannot start background task: no valid credentials."
started: "🔄 Background task #{number} started: \"{preview}\""
task_id: "Task ID: {task_id}"
keep_chatting: "You can continue chatting — results will appear when done."
label: "Background task #{number}"
complete: "✅ Background task #{number} complete"
prompt: "Prompt: \"{preview}\""
title_suffix: "(background #{number})"
no_response: "(No response generated)"
side_result:
complete: "{label} complete"
failed: "❌ {label} failed: {error}"
login:
label: "Sign-in"
title_suffix: "(sign-in)"
no_result: "(No result)"
btw:
usage: "Usage: /btw <question>\nExample: /btw which file was that error in?\nAnswers a quick question about this conversation without interrupting it.\n(For an independent background task, use /bg <prompt>.)"
no_credentials: "(>_<) Cannot answer side question: no valid credentials."
header: "💬 /btw: \"{preview}\""
title_suffix: "(btw)"
no_answer: "(No answer generated)"
bundles:
unavailable: "Bundle subsystem unavailable: {error}"
none_installed: "No skill bundles installed."
create_hint: "Create one with: hermes bundles create <name> --skill <s1> --skill <s2>"
directory: "Directory: {dir}"
title: "Skill Bundles"
installed_count: "({count} installed):"
default_desc: "Load {count} skills"
skills_count: "({count} skills)"
invoke_hint: "Invoke a bundle with /<slug>. Manage with `hermes bundles`."
browser:
engine_not_in_use: "⚠ browser.engine is 'lightpanda' but it is NOT in use: {reason}"
engine_lightpanda: "Engine: Lightpanda — {reason} (no screenshots)"
binary: "Binary: {path}"
binary_not_found: "⚠ lightpanda binary not found — {hint}"
use_usage: "Usage: /browser use [off]"
use_on_hint: "/browser use — switch to Browser Use mode (browser_exec via CLI 3.0)"
use_off_hint: "/browser use off — revert to the built-in browser tools"
use_enabled: "🌐 Browser Use mode enabled — browser_exec via the Browser Use CLI 3.0"
use_disabled: "🌐 Browser Use mode disabled — built-in browser tools restored"
session_reset: "Session reset. New tool configuration is active."
scheme_missing: "(missing)"
unsupported_scheme: "⚠ Unsupported browser url scheme: {scheme} (expected one of: http, https, ws, wss)"
invalid_port: "⚠ Invalid port in browser url: {url}"
missing_host: "⚠ Missing host in browser url: {url}"
port_occupied: "⚠ Port {port} is occupied by another application that isn't a CDP browser"
port_occupied_detail: "(an IDE debugger or dev server may be using it) — launching on port {launch_port} instead..."
launching: "Chromium-family browser isn't running with remote debugging — attempting to launch..."
launch_failed: "⚠ Could not auto-launch a Chromium-family browser"
launch_manually: "Launch a Chromium-family browser manually:"
no_executable: "No supported Chromium-family browser executable found in this environment"
launched: "✓ Chromium-family browser launched and listening on port {port}"
not_responding: "⚠ Browser launched but port {port} isn't responding yet"
try_again: "Try again in a few seconds — the debug instance may still be starting"
already_listening: "✓ Chromium-family browser is already listening at {url}"
port_unreachable: "⚠ Port {port} is not reachable at {url}"
not_connected: "Browser not connected — start a Chromium-family browser with remote debugging and retry /browser connect"
connected: "🌐 Browser connected to live Chromium-family browser via CDP"
endpoint: "Endpoint: {url}"
already_default: "Browser is not connected to a live Chromium-family browser (already using default mode)"
disconnected: "🌐 Browser disconnected from live Chromium-family browser"
reverted: "Browser tools reverted to default mode (local headless or cloud provider)"
status_lightpanda: "🌐 Browser: local Lightpanda (agent-browser --engine lightpanda)"
status_lightpanda_note: "⚡ Lightpanda: faster navigation, no screenshot support"
status_lightpanda_fallback: "Automatic Chromium fallback for screenshots and failed commands"
status_chrome: "🌐 Browser: local headless Chromium (agent-browser --engine chrome)"
status_auto: "🌐 Browser: local headless Chromium (agent-browser)"
status_browser_use: "🌐 Browser: Browser Use mode (browser_exec via the Browser Use CLI 3.0)"
status_browser_use_note: "Local Chrome via CDP, or Browser Use cloud browsers"
status_use_off_hint: "/browser use off — revert to the built-in browser tools"
status_connected: "🌐 Browser: connected to live Chromium-family browser via CDP"
status_reachable: "Status: ✓ reachable"
status_unreachable: "Status: ⚠ not reachable (browser may not be running)"
status_cloud: "🌐 Browser: {provider} (cloud)"
status_connect_hint: "/browser connect — connect to your live Chromium-family browser"
status_disconnect_hint: "/browser disconnect — revert to default"
usage: "Usage: /browser connect|disconnect|status|use"
usage_connect: "connect Connect browser tools to your live Chromium-family browser session"
usage_disconnect: "disconnect Revert to default browser backend"
usage_status: "status Show current browser mode"
usage_use: "use [off] Switch to Browser Use mode (CLI 3.0) / back to built-in tools"
heartbeat:
label: "Heartbeats"
paused: "⏸ Heartbeat paused: {prompt}"
none_set: "No heartbeat set."
none_to_resume: "No heartbeat to resume."
resumed: "▶ Heartbeat resumed (every {interval}): {prompt}"
cleared: "✓ Heartbeat cleared."
usage: "Usage: /heartbeat every <interval> <prompt> (e.g. /heartbeat every 10m Check CI)"
usage_also: "Also: /heartbeat status | pause | resume | clear"
interval_too_small: "Interval too small — minimum is {min_seconds}s."
prompt_required: "Usage: /heartbeat every <interval> <prompt> — the prompt is required."
invalid_label: "Invalid heartbeat"
set: "♥ Heartbeat set (every {interval}): {prompt}"
set_note: "Fires as a normal turn whenever the session is idle and the interval has elapsed. /heartbeat pause | resume | clear to manage; lives only while this Hermes process runs — use `hermes cron` for durable schedules."
refine:
nothing_yet: "Nothing to refine yet — send a message first."
empty: "Nothing to refine yet — the conversation is empty."
failed: "/refine failed to start: {error}"
focus_suffix: " (focus: {focus})"
started: "⚗ Reviewing this conversation in the background{focus} — any memory/skill updates will be reported when done."
review:
nothing_yet: "Nothing to review yet — send a message first."
failed: "/review failed to start: {error}"
goal:
label: "Goals"
continuing: "Continuing now — taking the next step."
send_to_kick: "Send any message to kick off the next step."
loop:
label: "Loops"
goal_defers_note: "Note: an active /goal is driving this session — loop wakeups defer until the goal finishes, pauses, or parks."
subgoal:
no_goal: "No active goal. Set one with /goal <text>."
usage_remove: "Usage: /subgoal remove <n>"
remove_needs_int: "/subgoal remove: <n> must be an integer (1-based index)."
removed: "✓ Removed subgoal {index}: {text}"
cleared_one: "✓ Cleared {count} subgoal."
cleared_other: "✓ Cleared {count} subgoals."
none_to_clear: "No subgoals to clear."
added: "✓ Added subgoal {index}: {text}"
skin:
unavailable: "Skin engine not available."
current: "Current skin: {name}"
available_header: "Available skins:"
usage: "Usage: /skin <name>"
custom_hint: "Custom skins: drop a YAML file in {dir}/skins/"
unknown: "Unknown skin: {name}"
available: "Available: {names}"
saved_suffix: " (saved)"
set_to: "Skin set to: {name}{saved}"
banner_note: "Note: banner colors will update on next session start."
colors_updated: "Prompt + TUI colors updated."
prompt_compose:
editor_header: "#! Compose your prompt below. Lines starting with '#!' are ignored.\n#! Save and quit to send; leave empty to cancel."
editor_failed: "(>_<) Could not open editor: {error}"
empty: "(._.) Empty prompt — nothing sent."
focus:
usage: "Usage: /focus [on|off|status]"
toggle:
save_failed: "Failed to save {setting} setting to config.yaml"
footer:
usage: "/footer [on|off|status]"
label: "Runtime footer"
fields: "Fields: {fields}"
timestamps:
usage: "/timestamps [on|off|status]"
label: "Message timestamps"
update:
managed_action: "update Hermes Agent"
choice_update: "Update Now"
choice_update_desc: "exit the current session and update Hermes Agent"
choice_cancel: "Cancel"
choice_cancel_desc: "keep the current session"
title: "☤ Update Hermes Agent"
detail: "This will exit the current session and run `hermes update`."
cancelled: "🟡 /update cancelled."
launching: "☤ Launching update..."
voice:
unknown_subcommand: "Unknown voice subcommand: {subcommand}"
usage: "Usage: /voice [on|off|tts|status]"
wake:
unknown_subcommand: "Unknown wake subcommand: {subcommand}"
usage: "Usage: /wake [on|off|status]"
enabled_in_config: "Wake word enabled in config (wake_word.enabled: true)."
disabled_in_config: "Wake word disabled in config (wake_word.enabled: false)."
tip_line: "✦ Tip: {tip}"
banner:
context_too_low: "⚠️ Context length is only {tokens} tokens — this is likely too low for agent use with tools."
context_minimum: "Hermes needs at least {tokens} tokens. Tool schemas + system prompt use a large fixed prefix."
fix_ollama: "Ollama fix: OLLAMA_CONTEXT_LENGTH={tokens} ollama serve"
fix_lm_studio: "LM Studio fix: Set context length in model settings → reload model"
fix_local_server: "Fix: start your server with at least {tokens_k}K context (llama.cpp: -c {tokens}), or set model.ollama_num_ctx in config.yaml to the window it really serves"
fix_config: "Fix: Set model.context_length in config.yaml, or increase your server's context setting"
hermes_models_not_agentic: "⚠ Nous Research Hermes 3 & 4 models are NOT agentic and are not designed for use with Hermes Agent."
hermes_models_lack_tools: "They lack tool-calling capabilities required for agent workflows. Consider using an agentic model (Claude, GPT, Gemini, DeepSeek, etc.)."
hermes_models_switch_hint: "Switch with: /model sonnet or /model gpt5"
project_skills_loaded: "◆ {count} project skill(s) loaded from this repo"
project_skills_untrusted: "◆ {count} project skill(s) found in {root} but not loaded — run `hermes skills trust` to enable them."
command:
unknown_nothing_sent: "Unknown command {command} — nothing was sent.{hint}"
did_you_mean_one: "Did you mean {suggestion}?"
type_help: "Type /help for the full list."
ambiguous: "Ambiguous command: {command}"
did_you_mean_many: "Did you mean: {candidates}?"
config:
title: "(^_^) Configuration"
loaded: "(loaded)"
not_found: "(not found)"
microsoft_entra_id: "Microsoft Entra ID"
api_key_not_set: "Not set!"
value_not_set: "not set"
section_model: "Model"
section_terminal: "Terminal"
section_agent: "Agent"
section_session: "Session"
label_model: "Model:"
label_base_url: "Base URL:"
label_api_key: "API Key:"
label_environment: "Environment:"
label_ssh_target: "SSH Target:"
label_working_dir: "Working Dir:"
label_timeout: "Timeout:"
label_max_turns: "Max Turns:"
label_toolsets: "Toolsets:"
label_verbose: "Verbose:"
label_started: "Started:"
label_config_file: "Config File:"
context:
header: "🧠 Context Usage — {model}"
compute_failed: "(._.) Could not compute context breakdown: {error}"
error:
# Chat-panel copy for a failed model request (hermes_cli/cli_chat_error_copy.py).
auth: "Your {provider} key was rejected. Run `hermes model` to re-enter it."
billing: "Your {provider} account is out of credit. Top up at the provider, or run /model to switch."
model_not_found: "'{model}' isn't available on {provider}. Run /model to pick a valid model."
rate_limit: "Rate limited by {provider}; wait a minute or /model to switch."
upstream_blocked: "A firewall/CDN in front of {provider} blocked the request (not your key). Set a User-Agent via extra_headers, or /model to switch."
overloaded: "{provider} is overloaded right now. Send /retry in a moment, or /model to switch."
server_error: "{provider} had an internal error. Send /retry in a moment, or /model to switch."
timeout: "{provider} did not answer in time. Send /retry, or /model to switch."
unknown: "The model request failed. Run /model to switch or `hermes doctor` to check the setup."
the_provider: "the provider"
the_current_model: "the current model"
with_details: "{lead}\nDetails: {details}"
agent_init_failed: "Hermes couldn't start the model connection: {error}. Your message was not sent. Run `hermes doctor` to check the setup, or /model to pick a different provider."
gateway:
starting: "Starting Hermes Gateway (messaging platforms)..."
gateway_status:
header: "(✿◠‿◠) Gateway Status"
platform_config_header: "Messaging Platform Configuration:"
enabled: "Enabled"
not_configured: "Not configured ({env_var})"
conversations_persist: "Conversations persist until /new or /reset."
to_start: "To start the gateway:"
config_file: "Configuration file: {path}"
load_error: "Error loading gateway config: {error}"
to_configure: "To configure the gateway:"
step_env_vars: "1. Set environment variables:"
step_config: "2. Or configure settings in {path}"
help:
header: "(^_^)? Available Commands"
section_session: "Session"
section_configuration: "Configuration"
section_tools_skills: "Tools & Skills"
section_info: "Info"
section_exit: "Exit"
section_context: "Context"
section_background_automation: "Background & Automation"
usage_suffix: "{description} (usage: /{command} {args})"
alias_suffix: "{description} (alias for /{command})"
skill_commands: "Skill Commands"
skill_bundles: "Skill Bundles"
quick_commands: "Quick Commands"
n_installed: "{count} installed"
n_configured: "{count} configured"
n_skills: "{count} skills"
load_n_skills: "Load {count} skills"
no_skill_commands: "No skill commands installed."
matching_query: "matching '{query}'"
to_list_them: "to list them"
filtered_by: "Filtered by '{query}' — run /help for the full list."
tip_line: "Tip: /help skills lists skill commands · /help <text> filters · Ctrl+P opens the command palette"
multiline_hint: "Multi-line: Ctrl+J, Alt+Enter, or \\\\+Enter for a new line"
draft_editor_hint: "Draft editor: Ctrl+G (Alt+G in VSCode/Cursor)"
attach_image_hint: "Attach image: /image {path} or start your prompt with a local image path"
paste_image_hint: "Paste image: Alt+V (or /paste)"
insights:
no_session_data: "No session data yet."
mcp_watch:
changed_reload_skipped: "🔄 MCP server config changed — reload skipped (auto-reload disabled)."
not_applied_run: "New settings are NOT applied yet. To apply them now, run:"
cache_note: "⚠️ Note: /reload-mcp rebuilds the tool set and invalidates the provider prompt cache (next message re-sends full input tokens)."
changed_reloading: "🔄 MCP server config changed — reloading connections..."
plugin:
command_error: "Plugin command error: {error}"
quick:
no_target: "Quick command '{command}' has no target defined"
unsupported_type: "Quick command '{command}' has unsupported type (supported: 'exec', 'alias')"
no_command: "Quick command '{command}' has no command defined"
no_output: "Command returned no output"
timed_out: "Quick command timed out (30s)"
error: "Quick command error: {error}"
reload_mcp:
title: "⚠️ /reload-mcp — Prompt cache invalidation warning"
detail: "Reloading MCP servers rebuilds the tool set for this session and\ninvalidates the provider prompt cache. The next message will\nre-send full input tokens (can be expensive on long-context or\nhigh-reasoning models)."
choice_once: "Approve Once"
choice_once_hint: "reload now"
choice_always: "Always Approve"
choice_always_hint: "reload now and silence this prompt permanently"
choice_cancel: "Cancel"
choice_cancel_hint: "leave MCP tools unchanged"
unchanged: "MCP tools unchanged."
always_msg: "🔒 Future /reload-mcp calls will run without confirmation."
once_verb: "reloading"
reloading: "🔄 Reloading MCP servers..."
reconnected: "♻️ Reconnected: {names}"
added: "➕ Added: {names}"
removed: "➖ Removed: {names}"
tools_available: "🔧 {tools} tool(s) available from {servers} server(s)"
agent_updated: "✅ Agent updated — {count} tool(s) available"
reload_skills:
reloading: "🔄 Reloading skills..."
total: "📚 {count} skill(s) available"
added_header: "➕ Added Skills:"
removed_header: "➖ Removed Skills:"
session_store:
unavailable_rich: "[bold yellow]⚠ Session store unavailable[/bold yellow] — this conversation will [bold]NOT be saved[/bold] and cannot be resumed later. Searching past sessions is also disabled.\n Reason: {reason}.\n {action}\n [dim]Details: {details}[/dim]"
unavailable_plain: "WARNING: Session store unavailable — this conversation will NOT be saved and cannot be resumed later. Reason: {reason}. {action}"
shell:
command_denied: "Command denied: {reason}"
flagged_as_dangerous: "flagged as dangerous"
exited: "! exited {code}"
skills:
bundle_load_failed: "Failed to load bundle for {command}"
loading_bundle: "Loading bundle: {name} ({count} skills)"
skipped_missing: "Skipped missing skills: {names}"
stacked_load_failed: "Failed to load stacked skills for {command}"
loading_stacked: "Loading {count} stacked skills: {names}"
loading_skill: "Loading skill: {name}"
skill_load_failed: "Failed to load skill for {command}"
tools:
none_available: "(;_;) No tools available"
header: "(^_^)/ Available Tools"
total: "Total: {count} tools ヽ(^o^)ノ"
toolsets:
header: "(^_^)b Available Toolsets"
currently_enabled_legend: "(*) = currently enabled"
tip_enable_all: "Tip: Use 'all' or '*' to enable all toolsets"
example: "Example: python cli.py --toolsets web,terminal"
usage:
unknown_subcommand: "Unknown /usage subcommand: {args}. Try /usage or /usage reset [--force]."
reset_wrong_provider: "Banked usage resets are only available on the openai-codex provider."
reset_switch_hint: "Switch with `/model` or `hermes auth` first."
checking_reset_credits: "⏳ Checking banked reset credits..."
codex_timeout: "❌ Timed out talking to the Codex backend — try again shortly."
no_api_calls: "(._.) No API calls made yet in this session."
header_session: "📊 Session Token Usage"
label_model: "Model:"
label_input_tokens: "Input tokens:"
label_output_tokens: "Output tokens:"
label_reasoning_subset: "↳ Reasoning (subset):"
label_prompt_tokens_total: "Prompt tokens (total):"
label_completion_tokens: "Completion tokens:"
label_total_tokens: "Total tokens:"
label_api_calls: "API calls:"
label_session_duration: "Session duration:"
label_current_context: "Current context:"
label_messages: "Messages:"
label_compressions: "Compressions:"
verbose:
label_off: "Tool progress: OFF"
detail_off: "silent mode, just the final response."
label_new: "Tool progress: NEW"
detail_new: "show each new tool (skip repeats)."
label_all: "Tool progress: ALL"
detail_all: "show every tool call."
label_verbose: "Tool progress: VERBOSE"
detail_verbose: "full args, results, and think blocks."
whoami:
you_label: "You:"
you_value: "cli (local terminal)"
user_label: "User:"
tier_label: "Tier:"
tier_value: "unrestricted"
slash_commands_label: "Slash commands:"
slash_commands_value: "all available"
worktree:
shallow_clone_keeping: "⚠ Shallow clone — cannot verify push state, keeping: {path}"
next_session_deepens: "The next `hermes -w` session deepens the clone and prunes merged worktrees automatically."
unpushed_keeping: "⚠ Worktree has unpushed commits, keeping: {path}"
clean_up_manually: "To clean up manually: git worktree remove --force {path}"
cleaned_up: "✓ Worktree cleaned up: {path}"
display:
# Tool verbs, spinner/progress text, completion rows and failure suffixes (agent/display.py).
# ``verb.<tool>`` is resolved at call time by get_tool_verb(); the connector joins verb + preview.
verb:
web_search: "Searching the web"
web_extract: "Reading"
browser_navigate: "Browsing"
browser_click: "Clicking"
browser_type: "Typing"
read_file: "Reading"
write_file: "Writing"
patch: "Editing"
search_files: "Searching files"
terminal: "Running"
execute_code: "Running code"
image_generate: "Generating image"
video_generate: "Generating video"
text_to_speech: "Generating speech"
vision_analyze: "Looking at the image"
session_search: "Searching past sessions"
skill_view: "Reading skill"
skills_list: "Listing skills"
skill_manage: "Updating skill"
delegate_task: "Delegating"
cronjob_manage: "Scheduling"
clarify: "Asking"
memory: "Updating memory"
todo_list: "Updating tasks"
verb_connector:
search: " for "
default: " "
bridge_generating:
tool_call: "a tool call"
tool_search: "a tool search"
tool_describe: "tool details"
status_phrase:
verb: "is {verb}"
using_tool: "is using {tool}"
preview:
shell_plus_one: "{first} + {count} command"
shell_plus_many: "{first} + {count} commands"
delegate_tasks: "{count} tasks: {goals}"
delegate_parallel_tasks: "{count} parallel tasks"
todo_verb_updating: "updating"
todo_verb_planning: "planning"
todo_reading: "reading task list"
todo_tasks: "{verb} {count} task(s)"
missing_old_text: "<missing old_text>"
send_message_to: "to {target}: \"{message}\""
session_search_recall: "recall: \"{query}\""
diff:
review_header: " ┊ review diff"
omitted_lines: "… omitted {count} diff line(s)"
omitted_files: " across {count} additional file(s)/section(s)"
# Comma-separated; a skin's spinner.thinking_verbs wins over this list.
thinking_verbs: "pondering,contemplating,musing,cogitating,ruminating,deliberating,mulling,reflecting,processing,reasoning,analyzing,computing,synthesizing,formulating,brainstorming"
spinner:
tool_line: " [tool] {message}"
done_line: " [done] {message}{elapsed}"
failure:
file_not_found: "File not found: {name}"
terminal_backend_unavailable: "terminal backend unavailable"
exit_code: " [exit {code}]"
memory_full: " [full]"
generic: " [error]"
generic_word: "error"
# Completion rows: "┊ {emoji} {verb:9} {detail}" -- verbs are padded to 9 cells by the renderer.
cute:
verb:
search: "search"
fetch: "fetch"
terminal: "$"
read: "read"
write: "write"
patch: "patch"
find: "find"
grep: "grep"
navigate: "navigate"
snapshot: "snapshot"
click: "click"
type: "type"
scroll: "scroll"
back: "back"
press: "press"
images: "images"
vision: "vision"
plan: "plan"
recall: "recall"
memory: "memory"
skills: "skills"
skill: "skill"
create: "create"
speak: "speak"
send: "send"
cron: "cron"
exec: "exec"
browser: "browser"
delegate: "delegate"
proc: "proc"
fetch_pages: "pages"
todo_progress: "{done}/{total} task(s)"
todo_reading: "reading tasks"
todo_update_progress: "update {done}/{total} ✓"
todo_update_count: "update {count} task(s)"
todo_count: "{count} task(s)"
cron_task_fallback: "task"
cron_create: "create {label}"
cron_listing: "listing"
delegate_parallel: "parallel"
process_list: "ls processes"
snapshot_full: "full"
snapshot_compact: "compact"
images_extracting: "extracting"
vision_analyzing_page: "analyzing page"
skills_list: "list {category}"
skills_all: "all"
scroll:
up: "up"
down: "down"
left: "left"
right: "right"
fallback_tool_name: "tool"
fallback_done: "done"
completed: "┊ ⚡ {tool} completed {duration}"
# Busy acks / heartbeats (agent/session_activity.py).
iteration_progress:
unbounded: "iteration {n}"
bounded: "iteration {n}/{max}"
# Self-improvement review summaries (agent/background_review.py).
review:
routing_fallback_warning: "⚠ auxiliary.background_review.provider='{task_provider}' (model '{task_model}') could not be routed ({error}); falling back to {provider}/{model}"
reasoning_effort_warning: "⚠ auxiliary.background_review.reasoning_effort='{effort}' has no effect while the review model is not the main model"
memory_op_line: "{label} {glyph} {preview}"
skill_patched_verbose: "📝 Skill '{name}' patched: \"{old}\" → \"{new}\""
skill_changed_verbose: "📝 Skill '{name}' {verb}: {description}"
skill_message_verbose: "📝 {message}"
skill_action: "Skill {action}"
label_updated: "{label} updated"
skill_result: "Skill '{name}' {verb}{path}"
skill_verb:
created: "created"
rewritten: "rewritten"
patched: "patched"
written: "written"
removed: "removed"
deleted: "deleted"
label:
skill: "Skill"
memory: "Memory"
user: "User profile"
summary_cli: " 💾 Self-improvement review: {summary}"
summary_callback: "💾 Self-improvement review: {summary}"
aux_failure_label: "background review"
notice:
uncompressed_context_overflow: "⚠️ Session context (~{tokens} tokens) exceeds the model context window (~{context_length} tokens) with compression disabled (compression.enabled: false). Use /compact to compress history or enable compression in config.yaml."
explainer:
# Turn-completion explanations and file-mutation footers (agent/turn_explainers.py). Appended to
# the human-facing reply after the transcript row is written; the model never sees them.
no_reply_prefix: "⚠️ No reply: "
empty_response: "{model} didn't produce a reply this time, even after retries. Send `continue` to try again, or switch models with /model."
exit:
all_retries_exhausted_no_response: "the model provider didn't answer after all retries. Send /retry, or switch models with /model."
partial_stream_recovery: "streaming stopped early and only a partial response was recovered. Send `continue` to resume from where it stopped."
fallback_prior_turn_content: "no new content was produced this turn; showing recovered prior context. Send `continue` to retry."
redirect_restart_limit_exceeded: "the request was cancelled by a new correction on every attempt, so the turn stopped instead of retrying forever. Your last correction is queued as the next message."
rebuilt_restart_limit_exceeded: "every provider in the fallback chain kept failing over, so the turn stopped instead of retrying forever. Send `continue` or switch provider."
budget_exhausted: "the per-turn iteration/cost budget was exhausted before a final answer. Send `continue` to keep going."
ollama_runtime_context_too_small: "the local model's context window was too small to finish. Increase the context size or use a larger model."
pending_tool_result: "the turn stopped while a tool result was still pending and the model produced no follow-up text. Send `continue` to let it summarize."
interrupted_during_api_call: "the request was interrupted mid-call before a reply was received. Send `continue` to retry."
max_iterations_reached: "the maximum tool-iteration limit was reached before a final answer. Send `continue` to keep going, or raise `max_iterations`."
error_near_max_iterations: "an error occurred near the iteration limit before a final answer. Check the tool output above, then send `continue`."
repeated_outer_errors: "the turn kept failing with repeated errors and was stopped early instead of retrying forever. Check the errors above, then send `continue` to retry."
persistence:
compression: "the turn was stopped because another process was compressing this session. Your message should already be saved — please send it again after compression completes."
compression_closed: "the turn was stopped because this session was rotated by context compression and its live continuation could not be adopted. The storage itself is healthy — refresh the client (or start a new turn) so it picks up the new session id, then send your message again."
turn_lease: "the turn was stopped because another Hermes process took over this session. Your reply was not saved — wait for the other process to finish, then send your message again."
session_row_missing: "this session was deleted from storage while the chat was still open, and Hermes could not recreate it, so this message was not saved. Send your message again; if it keeps failing, run `hermes {profile_arg}doctor`."
locked: "the turn was stopped because session storage was busy (another Hermes process was writing to the state database). Your message should already be saved — please send it again in a moment."
replaced: "the session database file was replaced while Hermes was running, so this message was not saved (a copy is kept in {home}/sessions/). Stop Hermes (`hermes {profile_arg}gateway stop`), run `hermes {profile_arg}doctor` — not `hermes {profile_arg}doctor --fix`, which would repair the wrong file in place — then start it again and send your message once more. Advanced recovery steps are in the log."
deleted_wal: "another Hermes process still holds an old copy of the session database's write-ahead log, so Hermes stopped writing to keep the file safe and this message was not saved (a copy is kept in {home}/sessions/). Nothing is lost. Quit every Hermes process on this profile (Desktop app, `hermes {profile_arg}gateway stop`, dashboard, cron), run `hermes {profile_arg}doctor` — it names any process still holding the log — then start Hermes again and send your message once more. Do not run `doctor --fix` or delete any state.db files while they run. Guide: {recovery_docs}"
corrupt: "the turn was stopped because the state database reported structural corruption (the transcript would have been lost on restart). Freeing disk space will not help. Recovery options:\n1. Run `hermes {profile_arg}doctor --fix`\n2. Stop the gateway, then recover with:\n hermes {profile_arg}sessions recover --source {db_path} --inspect-only\n (if it reports recoverable) hermes {profile_arg}sessions recover --source {db_path} --output recovered-state.db\n — recovery snapshots the damaged file first; do NOT run `sqlite3 ... \".recover\"` against the live state.db, a vulnerable sqlite3 CLI can corrupt it further\n3. Restore from a backup in {backups_dir}/\nThen send your message again."
fts_index: "the turn was stopped because the session search index (FTS5) is corrupt and could not be detached, so this message was not saved. The message store itself is not damaged: do not run recovery tools or restore a backup. Run `hermes {profile_arg}doctor --fix` (or restart Hermes, which repairs the index on open), then send your message again."
disk: "Hermes couldn't save this conversation to disk, so it stopped rather than lose your messages. The disk is probably full: free some space (or fix the permissions on {home}/state.db), then send your message again."
default: "Hermes couldn't save this conversation, so it stopped rather than lose your messages. Possible causes: the drive is out of room, or another Hermes process is holding the database. Close other Hermes windows, run `hermes {profile_arg}doctor` to check storage, then send your message again."
file_mutation:
header: "⚠️ File-mutation verifier: {count} file edit(s) FAILED this turn despite any wording above that may suggest otherwise. Run `git status` or `read_file` to confirm what actually landed."
entry: " • `{path}` — [{tool}] {preview}"
failed: "failed"
shared:
and_more: " • … and {count} more"
the_model: "The model"
tips:
# CLI startup / /new discovery tips (hermes_cli/tips.py). Zero-padded index keeps YAML
# order stable; pickers read tips.tNNN for the active language.
t001: "/bg <prompt> runs a task in a separate session while your current one stays free; /btw <question> answers a side question about this conversation without interrupting it."
t002: "/branch forks the current session so you can explore a different direction without losing progress."
t003: "/compress manually compresses conversation context when things get long."
t004: "/rollback lists filesystem checkpoints — restore files the agent modified to any prior state."
t005: "/rollback diff 2 previews what changed since checkpoint 2 without restoring anything."
t006: "/rollback 2 src/file.py restores a single file from a specific checkpoint."
t007: "/title \"my project\" names your session — resume it later with /resume or hermes -c."
t008: "/resume picks up where you left off in a previously named session."
t009: "/queue <prompt> queues a message for the next turn without interrupting the current one."
t010: "/undo removes the last user/assistant exchange from the conversation."
t011: "/retry resends your last message — useful when the agent's response wasn't quite right."
t012: "/verbose cycles tool progress display: off → new → all → verbose."
t013: "/reasoning high increases the model's thinking depth. /reasoning show displays the reasoning."
t014: "/fast toggles priority processing for faster API responses (provider-dependent)."
t015: "/yolo skips all dangerous command approval prompts for the rest of the session."
t016: "/model lets you switch models mid-session — try /model sonnet or /model gpt-5."
t017: "/model --global changes your default model permanently."
t018: "/personality pirate sets a fun personality — 14 built-in options from kawaii to shakespeare."
t019: "/skin changes the CLI theme — try ares, mono, slate, poseidon, or charizard."
t020: "/statusbar toggles a persistent bar showing model, tokens, context fill %, cost, and duration."
t021: "/tools disable browser temporarily removes browser tools for the current session."
t022: "/browser connect attaches browser tools to your running Chromium-family browser via CDP."
t023: "/plugins lists installed plugins and their status."
t024: "/cron manages scheduled tasks — set up recurring prompts with delivery to any platform."
t025: "/reload-mcp hot-reloads MCP server configuration without restarting."
t026: "/usage shows token usage, cost breakdown, and session duration."
t027: "/insights shows usage analytics for the last 30 days."
t028: "/paste checks your clipboard for an image and attaches it to your next message."
t029: "/profile shows which profile is active and its home directory."
t030: "/config shows your current configuration at a glance."
t031: "/stop kills all running background processes spawned by the agent."
t032: "@file:path/to/file.py injects file contents directly into your message."
t033: "@file:main.py:10-50 injects only lines 10-50 of a file."
t034: "@folder:src/ injects a directory tree listing."
t035: "@diff injects your unstaged git changes into the message."
t036: "@staged injects your staged git changes (git diff --staged)."
t037: "@git:5 injects the last 5 commits with full patches."
t038: "@url:https://example.com fetches and injects a web page's content."
t039: "Typing @ triggers filesystem path completion — navigate to any file interactively."
t040: "Combine multiple references: \"Review @file:main.py and @file:test.py for consistency.\""
t041: "Alt+Enter inserts a newline for multi-line input. (Windows Terminal intercepts Alt+Enter — use Ctrl+Enter instead.)"
t042: "Ctrl+C interrupts the agent. Double-press within 2 seconds to force exit."
t043: "Ctrl+Z suspends Hermes to the background — run fg in your shell to resume."
t044: "Tab accepts auto-suggestion ghost text or autocompletes slash commands."
t045: "Type a new message while the agent is working to interrupt and redirect it."
t046: "Alt+V pastes an image from your clipboard into the conversation."
t047: "Pasting 5+ lines auto-saves to a file and inserts a compact reference instead."
t048: "hermes -c resumes your most recent CLI session. hermes -c \"project name\" resumes by title."
t049: "hermes -w creates an isolated git worktree — perfect for parallel agent workflows."
t050: "hermes -w -q \"Fix issue #42\" combines worktree isolation with a one-shot query."
t051: "hermes chat -t web,terminal enables only specific toolsets for a focused session."
t052: "hermes chat -s github-pr-workflow preloads a skill at launch."
t053: "hermes chat -q \"query\" runs a single non-interactive query and exits."
t054: "hermes chat --max-turns 1000 overrides the default 500-iteration limit per turn."
t055: "hermes chat --checkpoints enables filesystem snapshots before every destructive file change."
t056: "hermes --yolo bypasses all dangerous command approval prompts for the entire session."
t057: "hermes chat --source telegram tags the session for filtering in hermes sessions list."
t058: "hermes -p work chat runs under a specific profile without changing your default."
t059: "hermes doctor --fix diagnoses and auto-repairs config and dependency issues."
t060: "hermes dump outputs a compact setup summary — great for bug reports."
t061: "hermes config set KEY VALUE auto-routes secrets to .env and everything else to config.yaml."
t062: "hermes config edit opens config.yaml in your default editor."
t063: "hermes config check scans for missing or stale configuration options."
t064: "hermes sessions browse opens an interactive session picker with search."
t065: "hermes sessions stats shows session counts by platform and database size."
t066: "hermes sessions prune --older-than 30 cleans up old sessions."
t067: "hermes skills search react --source skills-sh searches the skills.sh public directory."
t068: "hermes skills check scans installed hub skills for upstream updates."
t069: "hermes skills tap add myorg/skills-repo adds a custom GitHub skill source."
t070: "hermes skills snapshot export setup.json exports your skill configuration for backup or sharing."
t071: "hermes mcp add github --command npx adds MCP servers from the command line."
t072: "hermes mcp serve runs Hermes itself as an MCP server for other agents."
t073: "hermes auth add lets you add multiple API keys for credential pool rotation."
t074: "hermes completion bash >> ~/.bashrc enables tab completion for all commands and profiles."
t075: "hermes logs -f follows agent.log in real time. --level WARNING --since 1h filters output."
t076: "hermes backup creates a zip backup of your entire Hermes home directory."
t077: "hermes profile create coder creates an isolated profile that becomes its own command."
t078: "hermes profile create work --clone copies your current config and keys to a new profile."
t079: "hermes update syncs new bundled skills to ALL profiles automatically."
t080: "hermes gateway install sets up Hermes as a system service (systemd/launchd)."
t081: "hermes memory setup lets you configure an external memory provider (Honcho, Mem0, etc.)."
t082: "hermes webhook subscribe creates event-driven webhook routes with HMAC validation."
t083: "Save money: hermes tools disables unused tools, hermes skills config trims skills down."
t084: "/reasoning low or /reasoning minimal cuts thinking depth below the default (medium) — faster, cheaper responses."
t085: "hermes models routes vision, compression, and aux tasks to cheaper models — cuts background token cost 85%+ without downgrading your main chat model."
t086: "Set display.bell_on_complete: true in config.yaml to hear a bell when long tasks finish."
t087: "Set display.streaming: true to see tokens appear in real time as the model generates."
t088: "Set display.show_reasoning: true to watch the model's chain-of-thought reasoning."
t089: "Set display.compact: true to reduce whitespace in output for denser information."
t090: "Set display.busy_input_mode: queue to queue messages instead of interrupting the agent, or steer to inject them mid-run via /steer."
t091: "Set display.resume_display: minimal to skip the full conversation recap on session resume."
t092: "Set compression.threshold: 0.50 to control when auto-compression fires (default: 50% of context)."
t093: "Set agent.max_turns: 1000 to let the agent take more tool-calling steps per turn."
t094: "Set file_read_max_chars: 200000 to increase the max content per read_file call."
t095: "Set approvals.mode: smart to let an LLM auto-approve safe commands and auto-deny dangerous ones."
t096: "Set fallback_model in config.yaml to automatically fail over to a backup provider."
t097: "Set privacy.redact_pii: true to hash user IDs and phone numbers before sending to the LLM."
t098: "Set browser.record_sessions: true to auto-record browser sessions as WebM videos."
t099: "Set worktree: true in config.yaml to always create a git worktree (same as hermes -w)."
t100: "Set security.website_blocklist.enabled: true to block specific domains from web tools."
t101: "Set cron.wrap_response: false to deliver raw agent output without the cron header/footer."
t102: "HERMES_TIMEZONE overrides the server timezone with any IANA timezone string."
t103: "Environment variable substitution works in config.yaml: use ${VAR_NAME} syntax."
t104: "Quick commands in config.yaml run shell commands instantly with zero token usage."
t105: "Custom personalities can be defined in config.yaml under agent.personalities."
t106: "provider_routing controls OpenRouter provider sorting, whitelisting, and blacklisting."
t107: "execute_code runs Python scripts that call Hermes tools programmatically — results stay out of context."
t108: "delegate_task spawns up to 10 concurrent sub-agents by default (delegation.max_concurrent_children) with isolated contexts for parallel work."
t109: "web_extract works on PDF URLs — pass any PDF link and it converts to markdown."
t110: "search_files is ripgrep-backed and faster than grep — use it instead of terminal grep."
t111: "patch uses 9 fuzzy matching strategies so minor whitespace differences won't break edits."
t112: "patch supports V4A format for bulk multi-file edits in a single call."
t113: "read_file suggests similar filenames when a file isn't found."
t114: "read_file auto-deduplicates — re-reading an unchanged file returns a lightweight stub."
t115: "browser_vision takes a screenshot and analyzes it with AI — works for CAPTCHAs and visual content."
t116: "browser_console can evaluate JavaScript expressions in the page context."
t117: "image_generate creates images with FLUX 2 Pro and automatic 2x upscaling."
t118: "text_to_speech converts text to audio — plays as voice bubbles on Telegram."
t119: "send_message can reach any connected messaging platform from within a session."
t120: "The todo tool helps the agent track complex multi-step tasks during a session."
t121: "session_search performs full-text search across ALL past conversations."
t122: "The agent automatically saves preferences, corrections, and environment facts to memory."
t123: "/moa routes one hard prompt through your configured Mixture of Agents model set."
t124: "Terminal commands support background mode with notify_on_complete for long-running tasks."
t125: "Terminal background processes support watch_patterns to alert on specific output lines."
t126: "The terminal tool supports 6 backends: local, Docker, SSH, Modal, Daytona, and Singularity."
t127: "Each profile gets its own config, API keys, memory, sessions, skills, and cron jobs."
t128: "Profile names become shell commands — 'hermes profile create coder' creates the 'coder' command."
t129: "hermes profile export coder -o backup.tar.gz creates a portable profile archive."
t130: "If two profiles accidentally share a bot token, the second gateway is blocked with a clear error."
t131: "Sessions auto-generate descriptive titles after the first exchange — no manual naming needed."
t132: "Session titles support lineage: \"my project\" → \"my project #2\" → \"my project #3\"."
t133: "When exiting, Hermes prints a resume command with session ID and stats."
t134: "hermes sessions export backup.jsonl exports all sessions for backup or analysis."
t135: "hermes -r SESSION_ID resumes any specific past session by its ID."
t136: "Memory is a frozen snapshot — changes appear in the system prompt only at next session start."
t137: "Memory entries are automatically scanned for prompt injection and exfiltration patterns."
t138: "The agent has two memory stores: personal notes (~2200 chars) and user profile (~1375 chars)."
t139: "Corrections you give the agent (\"no, do it this way\") are often auto-saved to memory."
t140: "Over 80 bundled skills covering github, creative, mlops, productivity, research, and more."
t141: "Every installed skill automatically becomes a slash command — type / to see them all."
t142: "hermes skills install official/security/1password installs optional skills from the repo."
t143: "Skills can restrict to specific OS platforms — some only load on macOS or Linux."
t144: "skills.external_dirs in config.yaml lets you load skills from custom directories."
t145: "The agent can create its own skills as procedural memory using skill_manage."
t146: "/plan writes a markdown implementation plan to .hermes/plans/ without executing anything."
t147: "Cron jobs can attach skills: hermes cron add --skill blogwatcher \"Check for new posts\"."
t148: "Cron delivery targets include telegram, discord, slack, email, sms, and 12+ more platforms."
t149: "If a cron response starts with [SILENT], delivery is suppressed — useful for monitoring-only jobs."
t150: "Cron supports relative delays (30m), intervals (every 2h), cron expressions, and ISO timestamps."
t151: "Cron jobs run in completely fresh agent sessions — prompts must be self-contained."
t152: "Voice mode works with zero API keys if faster-whisper is installed (free local speech-to-text)."
t153: "Five TTS providers available: Edge TTS (free), ElevenLabs, OpenAI, NeuTTS (free local), MiniMax."
t154: "/voice on enables voice mode in the CLI. Ctrl+B toggles push-to-talk recording."
t155: "Streaming TTS plays sentences as they generate — you don't wait for the full response."
t156: "Voice messages on Telegram, Discord, WhatsApp, and Slack are auto-transcribed."
t157: "Hermes runs on 21 messaging platforms: Telegram, Discord, Slack, WhatsApp, Signal, Matrix, IRC, Microsoft Teams, email, and more."
t158: "hermes gateway install sets it up as a system service that starts on boot."
t159: "DingTalk uses Stream Mode — no webhooks or public URL needed."
t160: "BlueBubbles brings iMessage to Hermes via a local macOS server."
t161: "Webhook routes support HMAC validation, rate limiting, and event filtering."
t162: "The API server exposes an OpenAI-compatible endpoint compatible with Open WebUI and LibreChat."
t163: "Discord voice channel mode: the bot joins VC, transcribes speech, and talks back."
t164: "group_sessions_per_user: true gives each person their own session in group chats."
t165: "/sethome marks a chat as the home channel for cron job deliveries."
t166: "The gateway supports inactivity-based timeouts — active agents can run indefinitely."
t167: "Dangerous command approval has 4 tiers: once, session, always (permanent allowlist), deny."
t168: "Smart approval mode uses an LLM to auto-approve safe commands and flag dangerous ones."
t169: "SSRF protection blocks private networks, loopback, link-local, and cloud metadata addresses."
t170: "Tirith pre-exec scanning detects homograph URL spoofing and pipe-to-interpreter patterns."
t171: "MCP subprocesses receive a filtered environment — only safe system vars pass through."
t172: "Context files (.hermes.md, AGENTS.md) are security-scanned for prompt injection before loading."
t173: "command_allowlist in config.yaml permanently approves specific shell command patterns."
t174: "Context auto-compresses when it reaches the threshold — memories are flushed and history summarized."
t175: "The status bar turns yellow, then orange, then red as context fills up."
t176: "SOUL.md is the agent's primary identity file — customize it to shape behavior."
t177: "Hermes loads project context from .hermes.md, AGENTS.md, CLAUDE.md, or .cursorrules (first match)."
t178: "Subdirectory AGENTS.md files are discovered progressively as the agent navigates into folders."
t179: "Context files are capped at 20,000 characters with smart head/tail truncation."
t180: "Five browser providers: local Chromium, Browserbase, Browser Use, Camofox, and Firecrawl."
t181: "Camofox is an anti-detection browser — Firefox fork with C++ fingerprint spoofing."
t182: "browser_navigate returns a page snapshot automatically — no need to call browser_snapshot after."
t183: "browser_vision with annotate=true overlays numbered labels on interactive elements."
t184: "hermes mcp opens an interactive picker of Nous-approved MCPs you can install in one keystroke."
t185: "hermes mcp catalog lists Nous-approved MCP servers shipped with the repo."
t186: "hermes mcp install <name> installs a catalog entry, prompts for credentials, and lets you pick which of its tools to enable."
t187: "MCP servers are configured in config.yaml — both stdio and HTTP transports supported."
t188: "Per-server tool filtering: tools.include whitelists and tools.exclude blacklists specific tools."
t189: "MCP servers auto-generate toolsets at runtime — hermes tools can toggle them per platform."
t190: "MCP OAuth support: auth: oauth enables browser-based authorization with PKCE."
t191: "Checkpoints have zero overhead when no files are modified — enabled by default."
t192: "A pre-rollback snapshot is saved automatically so you can undo the undo."
t193: "/rollback also undoes the conversation turn, so the agent doesn't remember rolled-back changes."
t194: "Checkpoints use shadow repos in ~/.hermes/checkpoints/ — your project's .git is never touched."
t195: "batch_runner.py processes hundreds of prompts in parallel for training data generation."
t196: "hermes chat -Q enables quiet mode for programmatic use — suppresses banner and spinner."
t197: "Trajectory saving (--save-trajectories) captures full tool-use traces for model training."
t198: "Three plugin types: general (tools/hooks), memory providers, and context engines."
t199: "hermes plugins install owner/repo installs plugins directly from GitHub."
t200: "External memory providers: Honcho, OpenViking, Mem0 bundled; Hindsight and more via hermes plugins install."
t201: "Plugin hooks include pre/post_tool_call, pre/post_llm_call, and transform_terminal_output for output canonicalization."
t202: "Prompt caching (Anthropic) reduces costs by reusing cached system prompt prefixes."
t203: "The agent auto-generates session titles in a background thread — zero latency impact."
t204: "Smart model routing can auto-route simple queries to a cheaper model."
t205: "Slash commands support prefix matching: /h resolves to /help, /mod to /model."
t206: "Dragging a file path into the terminal auto-attaches images or sends as context."
t207: ".worktreeinclude in your repo root lists gitignored files to copy into worktrees."
t208: "hermes acp runs Hermes as an ACP server for VS Code, Zed, and JetBrains integration."
t209: "Custom providers: save named endpoints in config.yaml under custom_providers."
t210: "HERMES_EPHEMERAL_SYSTEM_PROMPT injects a system prompt that's never persisted to history."
t211: "credential_pool_strategies supports fill_first, round_robin, least_used, and random rotation."
t212: "hermes auth add nous or hermes auth add openai-codex sets up OAuth-based providers."
t213: "The API server supports both Chat Completions and Responses API with server-side state."
t214: "tool_preview_length: 0 in config shows full file paths in the spinner's activity feed."
t215: "hermes status --deep runs deeper diagnostic checks across all components."
t216: "Cron jobs can attach a Python script (--script) whose stdout is injected into the prompt as context."
t217: "Cron scripts live in ~/.hermes/scripts/ and run before the agent — perfect for data collection pipelines."
t218: "prefill_messages_file in config.yaml injects few-shot examples into every API call, never saved to history."
t219: "SOUL.md completely replaces the agent's default identity — rewrite it to make Hermes your own."
t220: "SOUL.md is auto-seeded with a default personality on first run. Edit it to customize."
t221: "/compress <focus topic> allocates 60-70% of the summary budget to your topic and aggressively trims the rest."
t222: "On second+ compression, the compressor updates the previous summary instead of starting from scratch."
t223: "Before a gateway session reset, Hermes auto-flushes important facts to memory in the background."
t224: "network.force_ipv4: true in config.yaml fixes hangs on servers with broken IPv6 — monkey-patches socket."
t225: "The terminal tool annotates common exit codes: grep returning 1 = 'No matches found (not an error)'."
t226: "Failed foreground terminal commands auto-retry up to 3 times with exponential backoff (2s, 4s, 8s)."
t227: "Bare sudo commands are auto-rewritten to pipe SUDO_PASSWORD from .env — no interactive prompt needed."
t228: "execute_code helpers require explicit imports: from hermes_tools import json_parse, shell_quote, retry."
t229: "execute_code's 7 sandbox tools (web_search, terminal, read/write/search/patch) use RPC — never enter context."
t230: "Reading the same file region 3+ times triggers a warning. At 4+, it's hard-blocked to prevent loops."
t231: "write_file and patch detect if a file was externally modified since the last read and warn about staleness."
t232: "V4A patch format supports Add File, Delete File, and Move File directives — not just Update."
t233: "MCP servers can request LLM completions back via sampling — the agent becomes a tool for the server."
t234: "MCP servers send notifications/tools/list_changed to trigger automatic tool re-registration without restart."
t235: "delegate_task with acp_command: 'claude' spawns Claude Code as a child agent from any platform."
t236: "Delegation has a heartbeat thread — child activity propagates to the parent, preventing gateway timeouts."
t237: "When a provider returns HTTP 402 (payment required), the auxiliary client auto-falls back to the next one."
t238: "agent.tool_use_enforcement steers models that describe actions instead of calling tools — auto for GPT/Codex."
t239: "agent.restart_after_turn_timeout lets in-flight turns finish before /restart enters stop(); restart_drain_timeout is only the force-interrupt budget once stop() begins."
t240: "agent.api_max_retries (default 3) controls how many times the agent retries a failed API call before surfacing the error — lower it for fast fallback."
t241: "The gateway caches AIAgent instances per session — destroying this cache breaks Anthropic prompt caching."
t242: "Any website can expose skills via /.well-known/skills/index.json — the skills hub discovers them automatically."
t243: "The skills audit log at ~/.hermes/skills/.hub/audit.log tracks every install and removal operation."
t244: "Stale git worktrees are auto-cleaned on startup: clean, fully-merged trees get pruned; dirty or unpushed work is always preserved."
t245: "Profiles scope Hermes state via HERMES_HOME; host tool subprocesses keep your real HOME unless terminal.home_mode is profile."
t246: "HERMES_HOME_MODE env var (octal, e.g. 0701) sets custom directory permissions for web server traversal."
t247: "Container mode: place .container-mode in HERMES_HOME and the host CLI auto-execs into the container."
t248: "Ctrl+C has 5 priority tiers: cancel recording → cancel prompts → cancel picker → interrupt agent → exit."
t249: "Every interrupt during an agent run is logged to ~/.hermes/interrupt_debug.log with timestamps."
t250: "BROWSER_CDP_URL connects browser tools to any running Chromium-family browser — accepts WebSocket, HTTP, or host:port."
t251: "BROWSERBASE_ADVANCED_STEALTH=true enables advanced anti-detection with custom Chromium (Scale Plan)."
t252: "The CLI auto-switches to compact mode in terminals narrower than 80 columns."
t253: "Quick commands support two types: exec (run shell command directly) and alias (redirect to another command)."
t254: "Per-task delegation model: delegation.model and delegation.provider in config route subagents to cheaper models."
t255: "delegation.reasoning_effort independently controls thinking depth for subagents."
t256: "display.platforms in config.yaml allows per-platform display overrides: {telegram: {tool_progress: all}}."
t257: "human_delay.mode in config simulates human typing speed — configurable min_ms/max_ms range."
t258: "Config version migrations run automatically on load — new config keys appear without manual intervention."
t259: "GPT and Codex models get special system prompt guidance for tool discipline and mandatory tool use."
t260: "Gemini models get tailored directives for absolute paths, parallel tool calls, and non-interactive commands."
t261: "context.engine in config.yaml can be set to a plugin name for alternative context management strategies."
t262: "Browser page snapshots over 15,000 characters are truncated or auto-summarized; the full snapshot is saved to cache/web for read_file paging."
t263: "The compressor does a cheap pre-pass: tool outputs over 200 chars are replaced with placeholders before the LLM runs."
t264: "When compression fails, further attempts are paused for 10 minutes to avoid API hammering."
t265: "Long dangerous commands (>70 chars) get a 'view' option in the approval prompt to see the full text first."
t266: "Audio level visualization shows ▁▂▃▄▅▆▇ bars during voice recording based on microphone RMS levels."
t267: "Profile names cannot collide with existing PATH binaries — 'hermes profile create ls' would be rejected."
t268: "hermes profile create backup --clone-all copies everything (config, keys, SOUL.md, memories, skills, sessions)."
t269: "The voice record key is configurable via voice.record_key in config.yaml — not just Ctrl+B."
t270: ".cursorrules and .cursor/rules/*.mdc files are auto-detected and loaded as project context."
t271: "Context files support 10+ prompt injection patterns — invisible Unicode, 'ignore instructions', exfil attempts."
t272: "GPT-5 and Codex use 'developer' role instead of 'system' in the message format."
t273: "Per-task auxiliary overrides: auxiliary.vision.provider, auxiliary.compression.model, etc. in config.yaml."
t274: "The auxiliary client treats 'main' as a provider alias — resolves to your actual primary provider + model."
t275: "hermes claw migrate --dry-run previews OpenClaw migration without writing anything."
t276: "File paths pasted with quotes or escaped spaces are handled automatically — no manual cleanup needed."
t277: "Slash commands never trigger the large-paste collapse — /command with big arguments works correctly."
t278: "In interrupt mode, slash commands typed during agent execution bypass interrupt logic and run immediately."
t279: "HERMES_DEV=1 bypasses container mode detection for local development."
t280: "Each MCP server gets its own toolset (mcp-servername) that can be toggled independently via hermes tools."
t281: "MCP ${ENV_VAR} placeholders in config are resolved at server spawn — including vars from ~/.hermes/.env."
t282: "Skills from trusted repos (NousResearch) get a 'trusted' security level; community skills get extra scanning."
t283: "The skills quarantine at ~/.hermes/skills/.hub/quarantine/ holds skills pending security review."
t284: "/steer <prompt> injects a note after the next tool call — nudge direction mid-task without interrupting."
t285: "/goal <text> sets a standing Ralph-loop objective — Hermes auto-continues turn after turn until a judge says done."
t286: "/snapshot create [label] saves a full state snapshot of Hermes config; /snapshot restore <id> reverts later."
t287: "/copy [N] copies the last assistant response to your clipboard, or the Nth-from-last with a number."
t288: "/redraw forces a full UI repaint, fixing terminal drift after tmux resize or mouse selection artifacts."
t289: "/agents (alias /tasks) shows active agents and running background tasks across the current session."
t290: "/footer toggles the gateway footer on final replies showing model, context %, and cwd."
t291: "/busy queue|steer|interrupt controls what pressing Enter does while Hermes is working."
t292: "/topic in Telegram DMs enables user-managed multi-session topic mode — /topic <id> restores past sessions inline."
t293: "/approve session|always runs a pending dangerous command with your chosen trust scope; /deny rejects it."
t294: "/restart gracefully restarts the gateway after draining active runs, then pings the requester when back up."
t295: "/kanban boards switch <slug> changes the active multi-project Kanban board from inside chat."
t296: "/reload reloads ~/.hermes/.env into the running session — pick up new API keys without restarting."
t297: "cronjob with no_agent=True runs a script on schedule and sends its stdout directly — zero tokens, zero LLM."
t298: "An empty cron script stdout means silent tick — nothing is delivered, perfect for threshold watchdogs."
t299: "HERMES_CRON_MAX_PARALLEL (default 4) caps how many cron jobs run per tick so bursts don't saturate your keys."
t300: "Gateway hooks live under ~/.hermes/hooks/<name>/ with HOOK.yaml + handler.py — handler must be named `handle`."
t301: "Hook events include gateway:startup, session:start, agent:step, and command:* wildcard subscriptions."
t302: "Drop a ~/.hermes/BOOT.md checklist and a gateway:startup hook runs it as a one-shot agent every boot."
t303: "hermes curator run --dry-run previews what the curator would archive or consolidate without mutating anything."
t304: "hermes curator pin <skill> hard-fences a skill against both auto-archival and the agent's skill_manage tool."
t305: "hermes curator rollback restores skills from a pre-run snapshot — backups live under skills/.curator_backups/."
t306: "hermes auth reset <provider> [target] clears cooldowns and exhaustion flags on a whole credential pool, or on one credential by index, id, or label."
t307: "credential_pool_strategies.<provider>: round_robin cycles keys evenly instead of the fill_first default."
t308: "use_gateway: true per-tool routes web, image, tts, or browser through your Nous subscription — no extra keys."
t309: "provider_routing.data_collection: deny excludes data-storing providers on OpenRouter."
t310: "provider_routing.require_parameters: true only routes to providers that support every param in your request."
t311: "HERMES_TUI_RESUME=1 auto-re-attaches to the most recent TUI session on launch — handy after SSH drops."
t312: "HERMES_TUI_THEME=light|dark|<hex> forces the TUI theme on terminals that don't set COLORFGBG."
t313: "Ctrl+G or Ctrl+X Ctrl+E in the TUI opens the input buffer in $EDITOR for long multi-line prompts."
t314: "The TUI renders LaTeX inline — $E=mc^2$ becomes Unicode math instead of raw TeX."
t315: "hermes dashboard launches a local web UI at 127.0.0.1:9119 — zero data leaves localhost."
t316: "hermes dashboard embeds the full Hermes TUI in your browser via xterm.js and a WebSocket PTY."
t317: "Drop a YAML in ~/.hermes/dashboard-themes/ with two palette colors to reskin the entire dashboard."
t318: "Dashboard plugins are drop-in: manifest.json + JS bundle in ~/.hermes/dashboard-plugins/ — no npm build required."
t319: "layoutVariant: cockpit in a dashboard theme adds a 260px left rail that plugins can populate via the sidebar slot."
t320: "display.tool_progress_command: true exposes /verbose on messaging platforms; it's CLI-only by default."
t321: "HERMES_BACKGROUND_NOTIFICATIONS=result only pings when background tasks finish (vs concise/all/error/off)."
t322: "HERMES_WRITE_SAFE_ROOT restricts write_file/patch to directory prefixes; multiple paths via os.pathsep (: or ;)."
t323: "HERMES_IGNORE_RULES skips auto-injection of AGENTS.md, SOUL.md, .cursorrules, memory, and preloaded skills."
t324: "HERMES_ACCEPT_HOOKS auto-approves unseen shell hooks declared in config.yaml without a TTY prompt."
t325: "auxiliary.goal_judge.model routes the /goal judge to a cheap fast model to keep loop cost near zero."
t326: "Checkpoints skip directories with more than 50,000 files to avoid slow git operations on massive monorepos."
t327: "tts.provider: piper runs 44-language local TTS on CPU — voices auto-download to ~/.hermes/cache/piper-voices/."
t328: "tts.providers.<name>.type: command wires any CLI TTS engine with {input_path} and {output_path} placeholders."
t329: "API_SERVER_ENABLED=true runs an OpenAI-compatible endpoint alongside the gateway for Open WebUI and LibreChat."
t330: "GATEWAY_PROXY_URL runs a split setup: platform I/O locally, agent work delegated to a remote API server."
t331: "MATRIX_DEVICE_ID pins a stable device ID for E2EE — without it, keys rotate every start and historic decrypt breaks."
t332: "TELEGRAM_WEBHOOK_SECRET is required whenever TELEGRAM_WEBHOOK_URL is set — generate with openssl rand -hex 32."
t333: "batch_runner.py --resume content-matches completed prompts by text so dataset reorders don't re-run finished work."
t334: "/new starts a fresh session in place (alias /reset) — fresh session ID, clean history, CLI stays open."
t335: "/clear wipes the terminal screen AND starts a new session — one shortcut for a visual reset."
t336: "/history prints the current conversation in-line without leaving the CLI — useful for a quick re-read."
t337: "/save writes the current conversation to disk without ending the session."
t338: "/status shows session info at a glance: ID, title, model, token usage, and elapsed time."
t339: "/image <path> attaches a local image file for your next prompt without pasting or drag-and-drop."
t340: "/platforms shows gateway and messaging-platform connection status right from inside chat."
t341: "/commands paginates the full slash-command + installed-skill list — useful on platforms without tab completion."
t342: "/toolsets lists every available toolset so you know what -t/--toolsets accepts."
t343: "/voice tts toggles TTS-only mode — agent replies out loud but you still type your prompts."
t344: "/reload-skills re-scans ~/.hermes/skills/ so drop-in skills appear without restarting the session."
t345: "/indicator kaomoji|emoji|unicode|ascii picks the TUI busy-indicator style shown during agent runs."
t346: "/debug uploads a support bundle (system info + logs) and returns shareable links — works in chat too."
t347: "hermes -z \"<prompt>\" is the purest one-shot: final answer on stdout, nothing else — ideal for piping in scripts."
t348: "hermes chat --pass-session-id injects the session ID into the system prompt so the agent can self-reference it."
t349: "hermes chat --image path/to/pic.png attaches a local image to a single -q query without a separate upload step."
t350: "hermes chat --ignore-user-config skips the active user config — reproducible bug reports and CI runs."
t351: "hermes chat --source tool tags programmatic chats so they don't clutter hermes sessions list."
t352: "hermes dump --show-keys includes redacted API key fingerprints for deeper support debugging."
t353: "hermes sessions rename <ID> \"new title\" renames any past session; hermes sessions delete <ID> removes one."
t354: "hermes import restores a full Hermes backup zip; session JSON/JSONL exports import from the dashboard Sessions page."
t355: "hermes fallback manages the fallback_model chain interactively — no hand-editing config.yaml."
t356: "hermes pairing rotates the DM pairing token — the first messager after rotation claims access to the bot."
t357: "hermes setup walks first-time users through provider, keys, and platform wiring in one interactive flow."
t358: "hermes status --deep runs the full health sweep across every component; plain hermes status is the quick view."
t359: "HERMES_AGENT_TIMEOUT=0 disables the gateway inactivity kill for a running agent — use for long research runs."
t360: "HERMES_ENABLE_PROJECT_PLUGINS=1 auto-loads repo-local plugins from ./.hermes/plugins/ — trust-gated by design."
t361: "HERMES_DISABLE_FILE_STATE_GUARD=1 turns off the 'file changed since you read it' guard on patch and write_file."
t362: "HERMES_ALLOW_PRIVATE_URLS=true lets web tools hit localhost and private networks — off by default in gateway mode."
t363: "HERMES_OPTIONAL_SKILLS=name1,name2 auto-installs extra optional-catalog skills on first run per profile."
t364: "HERMES_BUNDLED_SKILLS points at a custom bundled-skill tree — used by Homebrew and Nix packaging."
t365: "HERMES_DUMP_REQUEST_STDOUT=1 dumps every API request payload to stdout instead of log files."
t366: "HERMES_OAUTH_TRACE=1 logs redacted OAuth token exchange and refresh attempts for debugging provider auth."
t367: "HERMES_STREAM_RETRIES (default 3) controls mid-stream reconnect attempts on transient network errors."
t368: "HERMES_GATEWAY_BUSY_ACK_ENABLED=false silences the ⚡/⏳/⏩ ack messages when a user messages a busy agent."
t369: "HERMES_AGENT_NOTIFY_INTERVAL (default 180s) sets how often the gateway pings with progress on long turns."
t370: "HERMES_RESTART_DRAIN_TIMEOUT (default 900s) caps how long /restart waits for in-flight runs before forcing."
t371: "HERMES_CHECKPOINT_TIMEOUT (default 30s) caps filesystem checkpoint creation — raise it on huge monorepos."
t372: "image_gen.model in config.yaml picks the FAL model: flux-2/klein, gpt-image-2, nano-banana-pro, and more."
t373: "image_gen.provider routes image generation through a plugin (OpenAI Images, Codex, FAL) instead of the default."
t374: "AUXILIARY_VISION_BASE_URL + AUXILIARY_VISION_API_KEY point vision analysis at any OpenAI-compatible endpoint."
t375: "security.tirith_fail_open: false makes Hermes block commands when the tirith scanner itself errors out."
t376: "TIRITH_FAIL_OPEN env var overrides the tirith_fail_open config — a quick toggle without editing config.yaml."
t377: "--source tool chats are excluded from hermes sessions list by default — set --source explicitly to see them."
t378: "Session IDs are timestamp-prefixed (20250305_091523_abcd) so sorting works naturally in ls and jq."
t379: "API_SERVER_MODEL_NAME customizes the model name on /v1/models — essential for multi-profile Open WebUI setups."
t380: "Dashboard plugins are served from /dashboard-plugins/<name>/ — drop files into ~/.hermes/dashboard-plugins/."
# Empty-composer example prompts (TUI). Generic on purpose: Hermes is not coding-only.
placeholder:
p01: "Ask anything, or type / for commands…"
p02: "Summarize what's in this folder"
p03: "Draft a reply to the last email in my inbox"
p04: "Plan a feature, then build it step by step"
p05: "Find and fix a failing test"
p06: "Research this topic and write me a brief"
p07: "What changed in this repo recently?"
p08: "Turn these notes into a to-do list"
p09: "Explain this error and how to fix it"
p10: "Set a reminder or schedule a recurring task"
p11: "Type / to browse commands, or Ctrl+P for the palette"